We live in the century of technology, where the enormous evolution of data and science has recently favored a strong interest in processing, transmitting, and storing information. If, in the past, only a human mind could extract... more
We live in the century of technology, where the enormous evolution of data and science has recently favored a strong interest in processing, transmitting, and storing information. If, in the past, only a human mind could extract meaningful information from image data, after decades of dedicated research, scientists have managed to build complex systems that can identify different areas, tables, and texts from scanned documents, all the obtained information being easily accessed and passed by one to another. Books, newspapers, maps, letters, drawings-all types of documents can be scanned and processed in order to become available in a digital format. In the digital world, the storage space is very small compared to physical documents, so these applications will replace millions of old paper volumes with a single memory disk and will be accessible at the same time for anyone using just Internet access and without having a risk of deterioration. Other problems, such as ecological issues, accessibility and flexibility constraints can be solved by the use of document image analysis systems. This article presents the methods and techniques used to process on-paper documents and convert them to electronic ones, starting from pixel level and getting to the level of the entire document. The main purpose of Document Image Analysis Systems is to recognize texts and graphical interpretations from images, extract, format and present their contained information accordingly to the people's needs. We will also try to provide solid ground for practitioners that implement systems from this category to enhance the unsupervised processing features in order to make physical documents easily available to the masses.
Research Interests:
We live in a highly technologized era, where a great number of physical documents have started or must be digitized in order to make them accessible anywhere, for a great number of people. Optical Character Recognition (OCR) is one of the... more
We live in a highly technologized era, where a great number of physical documents have started or must be digitized in order to make them accessible anywhere, for a great number of people. Optical Character Recognition (OCR) is one of the techniques which are widely used in order to recognize characters from specific images obtained after scanning. Different types of systems have been developed in order to perform Optical Character Recognition for various types of documents, but the task is not easy, as documents differ not in terms of content, but have also in formats, fonts, age or deterioration. After reviewing the existing systems, the paper at hand proposes one which uses two well-known OCR engines and a voting principle based on weights. There are also analyzed the results of our combined technique, as opposed to each individual approach of the two chosen engines.
Research Interests:
Technology advances to make life easier for people. We tend to surround us with devices as small as possible and with the highest computing power. The need for data access from everywhere is an important detail. As a consequence, digital... more
Technology advances to make life easier for people. We tend to surround us with devices as small as possible and with the highest computing power. The need for data access from everywhere is an important detail. As a consequence, digital documents have been gaining ground on printed ones and for some sectors, the latter were even replaced. The need and the obligation to preserve the written cultural heritage, represented by books and valuable documents, some of them rare and even unique, forced us to imagine a system that protects the patrimony but makes it also accessible. In order to make books easily available to the public and at the lowest possible risk for the protection of the originals, we came to the idea of designing and creating an efficient digitization system of these records. The current article presents the proposed architecture of a Document Image Analysis System that will process the information with individual modules for each type of operation. The main scope for such tool is to recognize information from the documents and extract them for electronic use. The flow of operations are indicated by user, some steps can be eliminated depending on the user's desire and needs. In order to design an efficient Document Image Analysis System, we need a 3 axis approach: Education-involving students that can receive tasks for replacing modules and validating their homework, Research-performing various tests and Performance-testing the module interconnection and enabling the system to be extremely configurable. No matter what axis is considered, the main scope is the flexibility of the system-performed by individual modules as physical binaries or collection of binaries that are linked via scripts. Each module is designed to accomplish a certain major task by executing several sub-tasks whose results, in most cases, are subject to an intelligent voting process that produces the module's output data.
Research Interests:
The main objective of the paper is to achieve the implementation of an economical and user-friendly neurofeedback tool that can be used to treat pain, addiction, aggression, anxiety, autism, depression, Schizophrenia, epilepsy, headaches,... more
The main objective of the paper is to achieve the implementation of an economical and user-friendly neurofeedback tool that can be used to treat pain, addiction, aggression, anxiety, autism, depression, Schizophrenia, epilepsy, headaches, insomnia, Tourette syndrome, Attention Deficit Hyperactivity Disorder (ADHD), also used for the treatment of brain damage caused by stroke, trauma, and other causes in developing countries using EEG signal analysis. The analysis and the detection of the EEG can be achieved with the help of a home-made neurofeedback-built system. The technical and clinical EEG spectrum can be subdivided into-Alpha, Beta, Theta and Gamma ranges. The report will focus on the frequencies of the Beta band in accordance to the analysis of the power values and the output of the EEG analysis will be required to control the video and-or audio feedback. [22]
The Cranial electrotherapy stimulation (CES) is a medical device that can be used to treat anxiety, depression, pain and insomnia in both adult and children by passing a form of electrical stimulation to the brain. Proposed method of... more
The Cranial electrotherapy stimulation (CES) is a medical device that can be used to treat anxiety, depression, pain and insomnia in both adult and children by passing a form of electrical stimulation to the brain. Proposed method of action, side-effects, safety factors, and treatment efficacy was put into consideration and it was confirmed that CES technology has equal or greater efficacy for the treatment of anxiety, depression, pain and insomnia compared to antidepressant medications with fewer side effects. The main purpose of the research is to implement a micro-current Arduino-based low cost user-friendly Cranial Electrotherapy Stimulation tool that can be used by adult and children in underprivileged or developing countries in order to give effective treatment for Pain, Depression, Insomnia and Anxiety by analyzing with the help of an in-house designed and built system. CES has been the subject of the investigation of analysts for quite a few years. CES is a strategy that helps someone to control his or her brain waves consciously. During CES treatment, the reading is carefully recorded in a real-time manner. At that point, the signals is recorded and display to the subjects by using online feedback loop method. CES reading is a totally non-intrusive system that can be connected over and again to patients and neuro-typical children and adults with moderately no hazard or constraint. CES is known as a corresponding and effective treatment of numerous brain dysfunctions but still has its own particular advantages and disadvantages in spite of the fact that it is a non-intrusive strategy, its legitimacy has been addressed regarding indisputable logical proof. The CES have been found to have less SMR and best amplitudes than neurotypical ones with a border of specific thresholds. The proposed low-cost accurate device is a promising tool for the treatment of Pain, Depression, Insomnia and Anxiety in adult and children with poor financial resources.
Technology advances to make life easier for people. We tend to surround us with devices as small as possible and with the highest computing power. The need for data access from everywhere is an important detail. As a consequence, digital... more
Technology advances to make life easier for people. We tend to surround us with devices as small as possible and with the highest computing power. The need for data access from everywhere is an important detail. As a consequence, digital documents have been gaining ground on printed ones and for some sectors, the latter were even replaced. The need and the obligation to preserve the written cultural heritage, represented by books and valuable documents, some of them rare and even unique, forced us to imagine a system that protects the patrimony but makes it also accessible. In order to make books easily available to the public and at the lowest possible risk for the protection of the originals, we came to the idea of designing and creating an efficient digitization system of these records. The current article presents the proposed architecture of a Document Image Analysis System that will process the information with individual modules for each type of operation. The main scope for such tool is to recognize information from the documents and extract them for electronic use. The flow of operations are indicated by user, some steps can be eliminated depending on the user's desire and needs. In order to design an efficient Document Image Analysis System, we need a 3 axis approach: Education-involving students that can receive tasks for replacing modules and validating their homework, Research-performing various tests and Performance-testing the module interconnection and enabling the system to be extremely configurable. No matter what axis is considered, the main scope is the flexibility of the system-performed by individual modules as physical binaries or collection of binaries that are linked via scripts. Each module is designed to accomplish a certain major task by executing several sub-tasks whose results, in most cases, are subject to an intelligent voting process that produces the module's output data.
Research Interests:
We live in the century of technology, where the enormous evolution of data and science has recently favored a strong interest in processing, transmitting, and storing information. If, in the past, only a human mind could extract... more
We live in the century of technology, where the enormous evolution of data and science has recently favored a strong interest in processing, transmitting, and storing information. If, in the past, only a human mind could extract meaningful information from image data, after decades of dedicated research, scientists have managed to build complex systems that can identify different areas, tables, and texts from scanned documents, all the obtained information being easily accessed and passed by one to another. Books, newspapers, maps, letters, drawings-all types of documents can be scanned and processed in order to become available in a digital format. In the digital world, the storage space is very small compared to physical documents, so these applications will replace millions of old paper volumes with a single memory disk and will be accessible at the same time for anyone using just Internet access and without having a risk of deterioration. Other problems, such as ecological issues, accessibility and flexibility constraints can be solved by the use of document image analysis systems. This article presents the methods and techniques used to process on-paper documents and convert them to electronic ones, starting from pixel level and getting to the level of the entire document. The main purpose of Document Image Analysis Systems is to recognize texts and graphical interpretations from images, extract, format and present their contained information accordingly to the people's needs. We will also try to provide solid ground for practitioners that implement systems from this category to enhance the unsupervised processing features in order to make physical documents easily available to the masses.
Research Interests:
Indoor positioning systems are becoming a required subsystem in many ambient assisted living scenarios. Another area that would greatly benefit from the enriched context of localization is the IoT (Internet of Things) device interaction.... more
Indoor positioning systems are becoming a required subsystem in many ambient assisted living scenarios. Another area that would greatly benefit from the enriched context of localization is the IoT (Internet of Things) device interaction. However, at this moment there aren't any satisfying technologies or approaches for precise indoor positioning. This paper proposes an indoor positioning method based on trilateration using Wi-Fi RSSI measurements with corrections to eliminate the noise and the bias. The proposed method is tested in a laboratory environment and results are described.
The opportunities offered by mobile technology continue to expand with the advances in wireless broadband. These days there are a lot of entrenched business solutions, upgraded for mobile devices, which likewise have direct use to... more
The opportunities offered by mobile technology continue to expand with the advances in wireless broadband. These days there are a lot of entrenched business solutions, upgraded for mobile devices, which likewise have direct use to services of the government services, work environment practices and involvement of stakeholders. Each advance has challenges. The usage of M-Government will bring several difficulties. Amid them, the most significant for M-Government is the inclusion of infrastructure advancement, protection of information, security, legitimate issues, mobile entrance rate, availability and compatibility. Mobile WiMAX technology is an innovative wireless technology intended to empower pervasive, rapid mobile Internet access to the vastest cluster of gadgets including notebook PCs, handsets, mobile phones and buyer electronics such as gaming gadgets, cameras, and more. Mobile WiMAX conveys cheap networks and it is the first of all Internet mobile IP solution empowering proficient and scalable networks for information, video and voice. This research concerns Mobile Government powered by Mobile WiMAX technology, which delivers the benefits for Mobile Government because of high capacity, reliability, security and all IP communications targeted for the specific environment of Iraq which pose specific challenges.
This paper was designed with the main purpose of analyzing the collaborative and independent teaching approaches and their results by comparing them with other learning methods used during the Software Project Management course at the... more
This paper was designed with the main purpose of analyzing the collaborative and independent teaching approaches and their results by comparing them with other learning methods used during the Software Project Management course at the "Politehnica" University of Bucharest. Based on the experience gathered with the two aforementioned approaches we tried a "mixed" anarchic-style learning method, which is also discussed in this paper.
This paper was designed with the main purpose of analyzing the competition and collaboration based learning approaches and their results by comparing them with other learning methods used during the Software Project Management course at... more
This paper was designed with the main purpose of analyzing the competition and collaboration based learning approaches and their results by comparing them with other learning methods used during the Software Project Management course at the "Politehnica" University of Bucharest. Based on the experience gathered with the two aforementioned approaches we tried a mixed competition-collaboration learning method, which is also discussed in this paper.
Starting with the 80's, technology capable of aiding damaged human senses started to emerge, the first commonly used device being the cochlear implant. However there are other types of devices, more or less sophisticated that are not well... more
Starting with the 80's, technology capable of aiding damaged human senses started to emerge, the first commonly used device being the cochlear implant. However there are other types of devices, more or less sophisticated that are not well known. This paper investigates existing technologies able to aid persons with a visual sense deficit, by converting the stimuli and sending them for processing to a different region of the brain or through a different path.
Research Interests:
Highlights on textured surfaces are linear combinations of diffuse and specular reflection components. It is sometimes necessary to separate these lighting elements or completely remove the specular light, especially as a preprocessing... more
Highlights on textured surfaces are linear combinations of diffuse and specular reflection components. It is sometimes necessary to separate these lighting elements or completely remove the specular light, especially as a preprocessing step for computer vision. Many methods have been proposed for separating the reflection components. The method presented in this article improves on an existing algorithm by porting it on the GPU in order to optimize the speed, using new features found in DirectX11. New test results are also offered.
Research Interests:
The purpose of the current paper is to present an approach that can possibly improve OCR results for highly degraded documents, and also minimize dictionary based OCR post-processing. The proposed method uses k-nearest neighbors'... more
The purpose of the current paper is to present an approach that can possibly improve OCR results for highly degraded documents, and also minimize dictionary based OCR post-processing. The proposed method uses k-nearest neighbors' classification to split the input image in homogeneous regions, based on font characteristics, and then applies normalized cross correlation between word regions to find possible duplicates of the same word. These word duplicates will be collected in word lists from which a single word in each list needs to be recognizable by the OCR. All the found fragments in a word list will be replaced by the fragment of the understood word and placed back in the image.
Research Interests:
This paper presents methods that can be used in generating an entire planet from mathematical objects, possibly starting from a small random seed. The planet will be generated only from specified mathematical objects, fractals, procedural... more
This paper presents methods that can be used in generating an entire planet from mathematical objects, possibly starting from a small random seed. The planet will be generated only from specified mathematical objects, fractals, procedural models and constructive solid geometry. This planet will include elements such as vegetation, mountains, water, waves, rocky and sandy soil and clouds.
This paper addresses the problem of Skew Detection and Correction. The problem is of critical importance in the automated content conversion systems domain, making libraries digitalization projects possible. The paper at hand consists in... more
This paper addresses the problem of Skew Detection and Correction. The problem is of critical importance in the automated content conversion systems domain, making libraries digitalization projects possible. The paper at hand consists in a comparison between the main types of skew detection algorithms and presents the reader with a study on their advantages and disadvantages as well as proposed improvements.
In this paper we would like to present and discuss a system that automatically classifies coins. This flexible system can identify coins having different features and being photographed in different light conditions. For this purpose a... more
In this paper we would like to present and discuss a system that automatically classifies coins. This flexible system can identify coins having different features and being photographed in different light conditions. For this purpose a set of strong techniques for thresholding, edge detection and frequency domain processing were used in order to generate a fingerprint as meaningful and as invariant as possible for every coin class. Usually, the capturing of digital images cannot be performed in best conditions and inconsistencies can arise due to various lighting conditions as well as the performance of the capturing device. This article is here to suggest a method to reduce problems generated by lighting, so that image characteristics are more accurate. The proposed solution improves an existing automatic coin classification algorithm by applying illumination correction before the actual classification.
One of the directions for paper document conservation is conversion to microfilms and scanned images. Since recently microfilming has been abandoned over digitization, there is a need for standards and guidelines for the conversion... more
One of the directions for paper document conservation is conversion to microfilms and scanned images. Since recently microfilming has been abandoned over digitization, there is a need for standards and guidelines for the conversion workflow. The article proposes a set of methodologies for calibrating scanning systems to ensure high quality reproduction of both microfilms and original paper prints, in terms of tonal reproduction, geometric distortion and image sharpness.
This paper presents a new conversion algorithm from color images to grayscale that attempts to overcome the drawbacks of computing the grayscale luminance value as a weighted sum of the linear-intensity values. The algorithm aims to... more
This paper presents a new conversion algorithm from color images to grayscale that attempts to overcome the drawbacks of computing the grayscale luminance value as a weighted sum of the linear-intensity values. The algorithm aims to optimize the difference between neighboring color pixels based on the "potential" luminance difference. The algorithm iteratively adjusts the values associated to each of the pixel values, so that eventually there is a relevant difference between adjacent pixels so that the features become more visible.
This paper describes a method for fast and reliable coin classification using multiple coin images, taken in various conditions by different people. It also presents a possible solution to the misclassifications when dealing with... more
This paper describes a method for fast and reliable coin classification using multiple coin images, taken in various conditions by different people. It also presents a possible solution to the misclassifications when dealing with illumination inconsistencies. It relies on automatically extracted edge and texture/color features. For edge detection, the Canny's edge detector algorithm is used, in a variant with dynamic threshold values, obtained using Otsu's threshold algorithm. In the end some experimental results are presented indicating that the proposed approach fulfills the given task effectively.
The following paper proposes a new method of transmitting video streaming data based on splitting the input video in multiframe chunks (cubes). These chunks are processed through an algorithm that applies operations of downscaling,... more
The following paper proposes a new method of transmitting video streaming data based on splitting the input video in multiframe chunks (cubes). These chunks are processed through an algorithm that applies operations of downscaling, upscaling and difference, resulting a pyramid of different resolution cubes and residues (hypercube). These residues are then transmitted to the client based on their importance. The method is adaptive to the fluctuations in the transmitting channel bandwidth and requires a minimum overhead over streaming actual video frames.
This paper will first show how range trees are represented and the basic operations that can be made on them. Using this information, some computational geometry problems will be presented where range trees can be used in order to reduce... more
This paper will first show how range trees are represented and the basic operations that can be made on them. Using this information, some computational geometry problems will be presented where range trees can be used in order to reduce the time complexity.
The aim of this paper is to offer a faster and more rigorous way for calculating the Beta-Shape. The Beta-Shape is a tight-fitting concave polygonal contour, similar to the Alpha Shape, but which remains connected when data points are... more
The aim of this paper is to offer a faster and more rigorous way for calculating the Beta-Shape. The Beta-Shape is a tight-fitting concave polygonal contour, similar to the Alpha Shape, but which remains connected when data points are farther away than the desired resolution. The proposed method uses a computational geometry approach. We also propose an improvement to the original Beta Shape, aimed at generating a more natural flowing geometry. The complexity of the original algorithm are calculated and compared to the complexity of this algorithm.
In an automatic document conversion system, which builds digital documents from scanned articles, there is the need to perform various adjustments before the scanned image is fed to the OCR system. This is because the OCR system is prone... more
In an automatic document conversion system, which builds digital documents from scanned articles, there is the need to perform various adjustments before the scanned image is fed to the OCR system. This is because the OCR system is prone to error when the text is not properly identified, aligned, de-noised, etc. Such an adjustment is the detection of page skew, an unintentional rotation of the page, which probably occurred during scanning. In this work we have used the Radon transform to detect the angle at which a page was skewed.
This project studies the possibility of developing a mobile robot that scans the environment using 3D sensors. The project also focuses on developing an algorithm that analyzes the 3D data received from the 3D sensors and runs an image... more
This project studies the possibility of developing a
mobile robot that scans the environment using 3D sensors. The
project also focuses on developing an algorithm that analyzes
the 3D data received from the 3D sensors and runs an image
stitching algorithm based on the features found in the images.
The goal is obtaining a large 3D panorama from all the
pictures that the 3D sensor has sent.
mobile robot that scans the environment using 3D sensors. The
project also focuses on developing an algorithm that analyzes
the 3D data received from the 3D sensors and runs an image
stitching algorithm based on the features found in the images.
The goal is obtaining a large 3D panorama from all the
pictures that the 3D sensor has sent.
In an automatic document conversion system, which builds digital documents from scanned articles, there is a need to perform various adjustments before the scanned image is fed to the layout analysis system. This is because the layout... more
In an automatic document conversion system, which builds digital documents from scanned articles, there is a need to perform various adjustments before the scanned image is fed to the layout analysis system. This is because the layout detection system is sensitive to errors when the page elements are not properly identified, represented, denoised, etc. Such an adjustment is the detection of foreground and background or simply called a document image binarization. This paper presents a new idea for treating the common problems which may occur during the binarization phase of the documents, which considers a parameter-free local binarization algorithm which dynamically computes the window size after it sets a threshold for the standard variation value of the window. This proved to offer consistent results for a wide variety of scanned documents consisting of various old newspapers and old library documents in different languages, both handwritten and textual documents.
The main goal of this paper is to present the architecture and functionality of e-Service based platform. The project is structured along several dimensions that follow the development of complementary services, integrated to support... more
The main goal of this paper is to present the architecture and functionality of e-Service based platform. The project is structured along several dimensions that follow the development of complementary services, integrated to support everyday work experience, research and learning in the University POLITEHNICA of Bucharest (UPB). The platform support is represented by Internet as a large scale distributed environment. The current evolution of the Internet can be viewed from multiple perspectives: service oriented (Internet of Services), users centered (Internet of People), real-world integration over Internet (Internet of Things), production and use of multimedia content over Internet. The main services in eUPB 1 platform are: (i) data retrieval, aggregation and search service, (ii) communication service for heterogeneous networks, (iii) mobile services to support context-aware applications, (iv) secure data delivery infrastructure for wireless sensor networks, (v) 3DUPB-the 3D MMO virtual replica of UPB, (vi) analysis and content extraction of scanned documents, and (vii) collaboration service. This is a position paper presenting the general architecture of eUPB and a description of each services device and functionality.
Skew detection and correction is an important step in automated content conversion systems, on which overall system performance is dependent. Although there are many working solutions at the present time, the search for an algorithm that... more
Skew detection and correction is an important step in automated content conversion systems, on which overall system performance is dependent. Although there are many working solutions at the present time, the search for an algorithm that can achieve good error rates in a fast running time and on different layout types is still open, so new solutions for skew detection are needed. The paper at hand presents a neighbor clustering based approach that has the classical advantages of this class of algorithms-the speed, but delivers better accuracy, comparable with that of Hough based solutions.
We present here the results of a 3 year long nearshoring experiment, conducted in a semi-controlled academic environment. The experiment involved teams of 10-15 people from 6 European universities and proved that the biggest issue in... more
We present here the results of a 3 year long nearshoring experiment, conducted in a semi-controlled academic environment. The experiment involved teams of 10-15 people from 6 European universities and proved that the biggest issue in nearshoring lies in the difference between the 2 involved teams in terms of technical background differences.
The testing of HL7 conformance has recently gained more and more popularity due to the importance of correct intercommunication between critical systems such as e-Health applications. The HL7 messages are very complex, and there are major... more
The testing of HL7 conformance has recently gained more and more popularity due to the importance of correct intercommunication between critical systems such as e-Health applications. The HL7 messages are very complex, and there are major differences between the versions of the HL7 protocol. This paper presents our testing solution, which offers a general and extensible way of addressing these issues and others. We used a standardized testing technology, TTCN-3, and its template matching mechanism to validate the messages exchanged by medical systems. The HL7 profile we tested was QED, but our method allows extension to any other profiles.
This paper aims to present an effective method for recognising the useful print space of high-quality digital images obtained by means of automatic scanning devices.The algorithm purports to become a valid technical solution for cropping... more
This paper aims to present an effective method for
recognising the useful print space of high-quality digital images
obtained by means of automatic scanning devices.The
algorithm purports to become a valid technical solution for
cropping both single-page images, and double-page ones – in
the latter case the gutter is detected and the final result is made
of two distinct images representing the left-hand-side page and
the right-hand-side page
recognising the useful print space of high-quality digital images
obtained by means of automatic scanning devices.The
algorithm purports to become a valid technical solution for
cropping both single-page images, and double-page ones – in
the latter case the gutter is detected and the final result is made
of two distinct images representing the left-hand-side page and
the right-hand-side page
The problem of converting scanned documents into electronic files, especially for large electronic libraries, has been intensely studied. Document image binarization is at the base of any conversion system and requires maximum quality for... more
The problem of converting scanned documents into
electronic files, especially for large electronic libraries, has
been intensely studied. Document image binarization is at the
base of any conversion system and requires maximum quality
for the output, since it affects all subsequent processing steps.
This paper proposes a local binarization algorithm which
dynamically computes the window size based on a decrease of
the standard deviation value within the window being enlarged.
This approach proved to offer more reliable results for a wide
variety of scans of old and degraded documents.
electronic files, especially for large electronic libraries, has
been intensely studied. Document image binarization is at the
base of any conversion system and requires maximum quality
for the output, since it affects all subsequent processing steps.
This paper proposes a local binarization algorithm which
dynamically computes the window size based on a decrease of
the standard deviation value within the window being enlarged.
This approach proved to offer more reliable results for a wide
variety of scans of old and degraded documents.
Virtual spaces are becoming more and more present as a medium of social interaction on the internet. The biggest challenge for a massively multiuser online (MMO) application is to be able to accommodate a large number of users while... more
Virtual spaces are becoming more and more present as a medium of social interaction on the internet. The biggest challenge for a massively multiuser online (MMO) application is to be able to accommodate a large number of users while maintaining a high degree of immersion for the users of the virtual space. Current 3D MMO Servers architectures try to solve this challenge by using traditional models such client-server or peer-to-peer but suffer from limitations regarding performance, fault tolerance and cost. In this paper we propose an innovative 3D MMO Server Architecture that uses GPGPU (General Purpose programming on Graphical Processing Units) in order to solve some of these limitations.
Skew detection and correction poses particular challenges in deformed images, as well as images suffering from camera lens distortions or distortions caused by non-flat scanning surfaces. The current paper introduces an algorithm that... more
Skew detection and correction poses particular challenges in deformed images, as well as images suffering from camera lens distortions or distortions caused by non-flat scanning surfaces. The current paper introduces an algorithm that yields good results on such images and proposes a confidence measurement to estimate the accuracy of the returned skew angle.
Over the past few years, computer assisted applications have become indispensable in many domains. The medical domain does not make an exception. The need for accuracy and speed has led to collaborations between programmers and doctors in... more
Over the past few years, computer assisted applications have become indispensable in many domains. The medical domain does not make an exception. The need for accuracy and speed has led to collaborations between programmers and doctors in improving the medical activities of gathering, processing and managing patient’s data, in giving diagnoses and even in performing surgeries.
We propose a technique for medical applications, meant to reconstruct the 3D shape of the human body parts (bones, muscular tissue, etc.), and to allow doctors to remodel the 3D shape (in order to simulate a prosthesis or to simulate the resulting shape of a body part after performing surgery).
We propose a technique for medical applications, meant to reconstruct the 3D shape of the human body parts (bones, muscular tissue, etc.), and to allow doctors to remodel the 3D shape (in order to simulate a prosthesis or to simulate the resulting shape of a body part after performing surgery).
This paper describes an approach towards an orientation and skew detection for texts in scanned documents. Before using OCR systems to obtain character information from images, a preprocessing stage, comprising a number of adjustments,... more
This paper describes an approach towards an orientation and skew detection for texts in scanned documents. Before using OCR systems to obtain character information from images, a preprocessing stage, comprising a number of adjustments, has to be performed in order to obtain accurate results. One important operation that has to be considered is the skew correction, or deskewing, of the image, a fault that arises from an incorrect scanning process. This paper presents an iterative method for detecting the text orientation and skew angle, method based on histogram processing.
Covering the need for standards and guidelines in the workflow of paper document conservation through scanning, the article proposes a methodology for ensuring high quality scans in terms of image sharpness and an improved algorithm for... more
Covering the need for standards and guidelines in the
workflow of paper document conservation through scanning,
the article proposes a methodology for ensuring high quality
scans in terms of image sharpness and an improved algorithm
for computing the Modulation Transfer Function (MTF) of a
scanning system using the slanted edge method.
workflow of paper document conservation through scanning,
the article proposes a methodology for ensuring high quality
scans in terms of image sharpness and an improved algorithm
for computing the Modulation Transfer Function (MTF) of a
scanning system using the slanted edge method.
This paper presents a new approach for black and white conversion of greyscale images. The algorithm can be easily modified to work on colour images by changing the comparison pattern of the peaks. It uses a scanning method for black... more
This paper presents a new approach for black and white conversion of greyscale images. The algorithm can be easily modified to work on colour images by changing the comparison pattern of the peaks. It uses a scanning method for black pixel peaks and an independent threshold associated to a histogram of the scanned area. The results have been good for different types of documents.
Partitioning is a central problem in VLSI design automation, addressing circuit’s manufacturability. Circuit partitioning has multiple applications in VLSI design. One of the most common is that of dividing combinational circuits (usually... more
Partitioning is a central problem in VLSI design automation, addressing circuit’s manufacturability. Circuit partitioning has multiple applications in VLSI design. One of the most common is that of dividing combinational circuits (usually large ones) that will not fit on a single package among a number of packages. Partitioning is of practical importance for k-LUT based FPGA circuit implementation. In this work is presented new multilevel and multi-resource partitioning algorithm targeting large combinational circuits in order to efficiently use existing FPGAs circuits
Today's computer 3D graphics simulations need to provide a high degree of realism in order to create a rich and credible 3D virtual environment. Ray Tracing is a rendering method that can achieve such a degree of realism compared to the... more
Today's computer 3D graphics simulations need to provide a high degree of realism in order to create a rich and credible 3D virtual environment. Ray Tracing is a rendering method that can achieve such a degree of realism compared to the rendering mechanisms used by current graphics hardware. Until recently, Ray Tracing was not suitable to be used as a real time renderer because it is a heavy computational method and the hardware was not fast enough. In this paper we are proposing a solution to run Ray Tracing as a Multi GPGPU (General-purpose computing on graphics processing units) program and see how well Ray Tracing scales when implemented on multiple GPUs.
The research problem presented in this paper concerns the recognition of the print space in scanned images obtained through the use of automatic scanning devices. The algorithm aims to become a reliable technical solution for one of the... more
The research problem presented in this paper
concerns the recognition of the print space in scanned images
obtained through the use of automatic scanning devices. The
algorithm aims to become a reliable technical solution for one
of the most critical stages in mass document digitization
projects. Large scale tests showed that the proposed method is
effective and practical.
concerns the recognition of the print space in scanned images
obtained through the use of automatic scanning devices. The
algorithm aims to become a reliable technical solution for one
of the most critical stages in mass document digitization
projects. Large scale tests showed that the proposed method is
effective and practical.
Document layout analysis is a process that attempts to break down the structure of a scanned document and extract important layout elements such as paragraphs, headlines, images and so on. After this process, there is a need to mark the... more
Document layout analysis is a process that attempts
to break down the structure of a scanned document and extract
important layout elements such as paragraphs, headlines,
images and so on. After this process, there is a need to mark the
identified elements through an encapsulating shape that is
visually clearly defined and easy to manipulate. This paper
describes a method for achieving this. The results are very
accurate and the method has a high degree of stability.
to break down the structure of a scanned document and extract
important layout elements such as paragraphs, headlines,
images and so on. After this process, there is a need to mark the
identified elements through an encapsulating shape that is
visually clearly defined and easy to manipulate. This paper
describes a method for achieving this. The results are very
accurate and the method has a high degree of stability.
The purpose of this paper is to present a model for separator detection using the Delaunay triangulation and a common sampling algorithm based on a triangular filter. By using a sampling algorithm, the blur effect is created and together... more
The purpose of this paper is to present a model for
separator detection using the Delaunay triangulation and a
common sampling algorithm based on a triangular filter. By
using a sampling algorithm, the blur effect is created and
together with the output triangles, obtained from the modified
Delaunay algorithm, the separators can be determined after
processing.
separator detection using the Delaunay triangulation and a
common sampling algorithm based on a triangular filter. By
using a sampling algorithm, the blur effect is created and
together with the output triangles, obtained from the modified
Delaunay algorithm, the separators can be determined after
processing.
Entity clustering is a vital feature of any automatic content conversion system. Such systems generate digital documents from hard copies of newspapers, books, etc. At application level, the system processes an image (usually in black and... more
Entity clustering is a vital feature of any automatic content conversion system. Such systems generate
digital documents from hard copies of newspapers, books, etc. At application level, the system processes an
image (usually in black and white color mode) and identifies the various content layout elements, such as
paragraphs, tables, images, columns, etc. Here is where the entity clustering mechanism comes into play. Its role
is to group atomic entities (characters, points, lines) into layout elements. To achieve this, the system takes on
different approaches which rely on the geometrical properties of the enclosed items: their relative position, size,
boundaries and alignment. This paper describes such an approach based on 3D mesh reduction.
digital documents from hard copies of newspapers, books, etc. At application level, the system processes an
image (usually in black and white color mode) and identifies the various content layout elements, such as
paragraphs, tables, images, columns, etc. Here is where the entity clustering mechanism comes into play. Its role
is to group atomic entities (characters, points, lines) into layout elements. To achieve this, the system takes on
different approaches which rely on the geometrical properties of the enclosed items: their relative position, size,
boundaries and alignment. This paper describes such an approach based on 3D mesh reduction.
This paper describes an approach towards obtaining the normalized measure of text resemblance in scanned images. The technique, aimed at automatic content conversion, is relying on the detection of standard character features and uses a... more
This paper describes an approach towards obtaining the normalized measure of text resemblance in
scanned images. The technique, aimed at automatic content conversion, is relying on the detection of standard
character features and uses a sequence of procedures and algorithms applied sequentially on the input document.
The approach makes use solely of the geometrical characteristics of characters, ignoring information regarding
context or the character-recognition.
scanned images. The technique, aimed at automatic content conversion, is relying on the detection of standard
character features and uses a sequence of procedures and algorithms applied sequentially on the input document.
The approach makes use solely of the geometrical characteristics of characters, ignoring information regarding
context or the character-recognition.
Bitonal conversion is a basic preprocessing step in Automatic Content Analysis, a very active research area in the past years. The information retrieval process is performed usually on black and white documents in order to increase... more
Bitonal conversion is a basic preprocessing step in Automatic Content Analysis, a very active
research area in the past years. The information retrieval process is performed usually on black and white
documents in order to increase efficiency and use simplified investigation techniques. This paper presents a
number of new modern conversion algorithms which are aimed at becoming an alternative to current
approaches used in the industry. The proposed methods are suitable for both scanned images and documents in
electronic format. Firstly, an algorithm consisting of a contrast enhancement step, followed by a conversion
based on adaptive levelling of the document is presented. Then a new multi-threshold technique is suggested as
a solution for noise interferences, a common feature of scanned books and newspapers. Finally, three more
approaches adapted to the particular properties of electronic documents are introduced. Experimental results are
given in order to verify the effectiveness of the proposed algorithms.
research area in the past years. The information retrieval process is performed usually on black and white
documents in order to increase efficiency and use simplified investigation techniques. This paper presents a
number of new modern conversion algorithms which are aimed at becoming an alternative to current
approaches used in the industry. The proposed methods are suitable for both scanned images and documents in
electronic format. Firstly, an algorithm consisting of a contrast enhancement step, followed by a conversion
based on adaptive levelling of the document is presented. Then a new multi-threshold technique is suggested as
a solution for noise interferences, a common feature of scanned books and newspapers. Finally, three more
approaches adapted to the particular properties of electronic documents are introduced. Experimental results are
given in order to verify the effectiveness of the proposed algorithms.
This paper describes a model for detecting all types of separators on a document page and combining the results towards obtaining elements clusters on every document image page. The separators are determined by using various methods as... more
This paper describes a model for detecting all types of separators on a document page and combining
the results towards obtaining elements clusters on every document image page. The separators are determined
by using various methods as for example the Delaunay triangulation. The physical layouts of the documents are
always hard to extract, but determining the simplest separators found in most documents is the starting point for
correct layout detection
the results towards obtaining elements clusters on every document image page. The separators are determined
by using various methods as for example the Delaunay triangulation. The physical layouts of the documents are
always hard to extract, but determining the simplest separators found in most documents is the starting point for
correct layout detection
This paper presents a new approach towards creating a type of hierarchy for document image page using the information given by the Delaunay triangulation. The steps of the algorithm are presented under the form of a cluster tree... more
This paper presents a new approach towards creating a type of hierarchy for document image page using the information given by the Delaunay triangulation. The steps of the algorithm are presented under the form of a cluster tree containing the information of the page in structures such as collections of pixels and using the distance between them as a binding measurement. The final result provides the page segmentation into clusters containing pictures, titles and paragraphs.
Existing Computational Geometry algorithms aren't able to create a tight-fitting contour around document elements. Some of them don't follow the data points close enough, generating overlapping elements, while others, trying to keep close... more
Existing Computational Geometry algorithms aren't able to create a tight-fitting contour around document elements. Some of them don't follow the data points close enough, generating overlapping elements, while others, trying to keep close to the contour, break one element into pieces, falling between white spaces. The presented method both follows the contour closely and generates a single shape for a single element. The generation of the shape can be stopped at any time, either when there are no more intersections between layout elements or after a certain time elapsed, the result being valid at any given moment. In association with other algorithms it can offer a fast and clean solution for the problem of finding the non-overlapping areas resulted from the Layout Analysis document processing phase.
Image document analysis is a complex process that involves several steps of processing. However, due to their sensitivity to errors, most of these are not applied on the original image; instead, they use a simplified black and white... more
Image document analysis is a complex process that involves several steps of processing. However, due to their sensitivity to errors, most of these are not applied on the original image; instead, they use a simplified black and white version of the original image, which offers a clear separation between foreground and background. Unfortunately, achieving the optimal separation is difficult to achieve, as no proposed algorithm has managed to offer a solution that is adequate for any type of input. This paper aims to present some of the more recent approaches in the field and compare their results with some of the classic algorithms.
Skew detection for scanned documents is an important field of research, of great interest for both academic and commercial environments. The Hough transform has been established as the standard for accurate skew detection, thanks to its... more
Skew detection for scanned documents is an important field of research, of great interest for both academic and commercial environments. The Hough transform has been established as the standard for accurate skew detection, thanks to its high accuracy and mathematical soundness. Unfortunately, this high accuracy comes with a high time penalty, and thus the search for speed improvements that do not reduce accuracy is on. The present article will present an impartial comparison of the proposed methods and point out each method strong and weak points.
This paper focuses on the applicability of ontologies in medical image processing area. Techniques for automatically describing medical images in a medical language that doctors can operate with are presented. Super resolution algorithms... more
This paper focuses on the applicability of ontologies in medical image processing area. Techniques for automatically describing medical images in a medical language that doctors can operate with are presented. Super resolution algorithms are also highlighted in order to prove their superiority over usual interpolation methods on medical images. At the end, a set of experimental results regarding ontologies are offered.
Identification of text lines in documents, or text line segmentation, represents the first step in the process called ‘Text recognition”, whose purpose is to extract the text and put it in a more understandable format. The paper proposes... more
Identification of text lines in documents, or text line segmentation, represents the first step in the process called ‘Text recognition”, whose purpose is to extract the text and put it in a more understandable format. The paper proposes a seam carving algorithm as an approach to find the text lines. This algorithm uses a new method that allocates dynamic weights for every processed pixel in the original image. With this addition, the resulting lines follow the text more accurately. The downside of this technique is the computational time overhead
When it comes to image segmentation, there is no single technique that can provide the best possible result for any type of image. Therefore, based on different approaches, numerous algorithms have been developed so far and each has its... more
When it comes to image segmentation, there is no single technique that can provide the best possible result for any type of image. Therefore, based on different approaches, numerous algorithms have been developed so far and each has its upsides and downsides, depending on the input data. This paper proposes a voting method that tries to merge different results of some well-known image segmentation algorithms into a relevant output, aimed to be, as frequently as possible, better than any of the independent ones previously computed.
The first step in the text recognition process is represented by the text line segmentation procedures. Only after text lines are correctly identified can the process proceed to the recognition of individual characters. This paper... more
The first step in the text recognition process is represented by the text
line segmentation procedures. Only after text lines are correctly identified can the
process proceed to the recognition of individual characters. This paper proposes a
line segmentation algorithm based on the computation of an information content level,
called energy, for each pixel of the image and using it to execute the seam carving
procedure. The algorithm proposes the identification of text lines which follow the
text more accurately with the expected downside of the computational overhead.
line segmentation procedures. Only after text lines are correctly identified can the
process proceed to the recognition of individual characters. This paper proposes a
line segmentation algorithm based on the computation of an information content level,
called energy, for each pixel of the image and using it to execute the seam carving
procedure. The algorithm proposes the identification of text lines which follow the
text more accurately with the expected downside of the computational overhead.
As a part of the Computer Vision domain, layout analysis is the process through which the regions of interest from a document available as an image are being classified. A scanned file could be an example of such a document. The... more
As a part of the Computer Vision domain, layout analysis is the process through which the regions of interest from a document available as an image are being classified. A scanned file could be an example of such a document. The components of the layout analysis process are: the geometrical analysis and the logical layout.
The geometrical analysis involves the detection and labeling of differing regions or blocks of the image as being text, illustration, mathematic symbols, tables etc.
The logical layout refers to the detecting of the logical role that various regions have in the document (titles, footnotes, etc.).
The layout analysis process is intended to be performed before the document is sent or the OCR engine but it can also be used to identify copies of the same document or for indexing documents by structure. This paper presents an approach to analyze the layout based on a voting scheme, thus combining a series of algorithms and using weighted, majority and unanimous votes, with the purpose of increasing the accuracy of the results.
The geometrical analysis involves the detection and labeling of differing regions or blocks of the image as being text, illustration, mathematic symbols, tables etc.
The logical layout refers to the detecting of the logical role that various regions have in the document (titles, footnotes, etc.).
The layout analysis process is intended to be performed before the document is sent or the OCR engine but it can also be used to identify copies of the same document or for indexing documents by structure. This paper presents an approach to analyze the layout based on a voting scheme, thus combining a series of algorithms and using weighted, majority and unanimous votes, with the purpose of increasing the accuracy of the results.
The process of document image preservation and analysis starts from a correct acquisition of scanned or photographed digital information. It may sound simple but, unfortunately, it is not. The photo or scanning devices employed in the... more
The process of document image preservation and
analysis starts from a correct acquisition of scanned or photographed
digital information. It may sound simple but, unfortunately, it is not.
The photo or scanning devices employed in the process need constant
check and maintenance in order to certify the quality of their output.
Even in case of perfectly functioning devices, small deviations may
appear. The purpose of this paper is to examine the available methods
of measuring the correctness of an imaging device functioning, to
propose a set of methodologies for specific target validation in terms
of tonal reproduction, geometric distortion, and image sharpness and
to ensure that the correct output is obtained if only minor deviations
occur.
analysis starts from a correct acquisition of scanned or photographed
digital information. It may sound simple but, unfortunately, it is not.
The photo or scanning devices employed in the process need constant
check and maintenance in order to certify the quality of their output.
Even in case of perfectly functioning devices, small deviations may
appear. The purpose of this paper is to examine the available methods
of measuring the correctness of an imaging device functioning, to
propose a set of methodologies for specific target validation in terms
of tonal reproduction, geometric distortion, and image sharpness and
to ensure that the correct output is obtained if only minor deviations
occur.
The following paper proposes a solution for the adaptive video streaming problem based on a construction of a pyramid of hypercubes. The hypercubes are built as residues between successive downsampling and upsampling operations over... more
The following paper proposes a solution for the
adaptive video streaming problem based on a construction of a
pyramid of hypercubes. The hypercubes are built as residues between
successive downsampling and upsampling operations over chunks of
video data. The described mechanism allows a great deal of
flexibility with minimum overhead, is both general and
mathematically elegant and may be further refined for better
transmission control.
adaptive video streaming problem based on a construction of a
pyramid of hypercubes. The hypercubes are built as residues between
successive downsampling and upsampling operations over chunks of
video data. The described mechanism allows a great deal of
flexibility with minimum overhead, is both general and
mathematically elegant and may be further refined for better
transmission control.
The Image Deblurring problem seems to be one of perpetual actuality. The research started in the Second World War for military purposes and continued until today proposing numerous techniques for a huge palette of applications, all aiming... more
The Image Deblurring problem seems to be one of
perpetual actuality. The research started in the Second World War for
military purposes and continued until today proposing numerous
techniques for a huge palette of applications, all aiming at recovering
the original signal from a blurred one. The current paper synthesizes
most of the efforts carried over the time, constructively comparing
available approaches and offering a high degree of up-to-date
completeness over this vast research subject.
perpetual actuality. The research started in the Second World War for
military purposes and continued until today proposing numerous
techniques for a huge palette of applications, all aiming at recovering
the original signal from a blurred one. The current paper synthesizes
most of the efforts carried over the time, constructively comparing
available approaches and offering a high degree of up-to-date
completeness over this vast research subject.
The purpose of this paper is to offer a solution for building a controllable in refinement, adaptive contour, in a robust, intuitive manner, and generalized for any input point set. The presented algorithm is both mathematically elegant... more
The purpose of this paper is to offer a solution for
building a controllable in refinement, adaptive contour, in a robust,
intuitive manner, and generalized for any input point set. The
presented algorithm is both mathematically elegant and general.
Despite the fact that the problem and the solution are discussed for
the two-dimensional case, the entire approach is valid for higher
dimensions as well.
building a controllable in refinement, adaptive contour, in a robust,
intuitive manner, and generalized for any input point set. The
presented algorithm is both mathematically elegant and general.
Despite the fact that the problem and the solution are discussed for
the two-dimensional case, the entire approach is valid for higher
dimensions as well.
This article proposes an alternate way for resolving classical computational geometry problems. The particularity is the integration of Range Tree data structures in solving problems like: segment intersections, orthogonal queries and... more
This article proposes an alternate way for resolving
classical computational geometry problems. The particularity is the
integration of Range Tree data structures in solving problems like:
segment intersections, orthogonal queries and calculation of
rectangular areas. For particular scenarios complexity improvements
can be observed. Given that these three algorithms were implemented
relying on range trees, the research opens the door for introducing
similar computational structures in related geometry problems.
classical computational geometry problems. The particularity is the
integration of Range Tree data structures in solving problems like:
segment intersections, orthogonal queries and calculation of
rectangular areas. For particular scenarios complexity improvements
can be observed. Given that these three algorithms were implemented
relying on range trees, the research opens the door for introducing
similar computational structures in related geometry problems.
This paper proposes a method of transmitting video streaming data based on downsampling-upsampling pyramidal decomposition. By implementing an octal tree decomposition of the frame cubes, prior to transforming them into hypercubes, the... more
This paper proposes a method of transmitting video streaming data based on
downsampling-upsampling pyramidal decomposition. By implementing an octal tree
decomposition of the frame cubes, prior to transforming them into hypercubes, the
algorithm manages to increase the granularity of the transmitted data. In this sense, the
communication relies on a series of smaller hypercubes, as opposed to a single hypercube
containing the entire, undivided frames form a sequence. This translates into increased
adaptability to the variations of the transmitting channel’s bandwidth.
downsampling-upsampling pyramidal decomposition. By implementing an octal tree
decomposition of the frame cubes, prior to transforming them into hypercubes, the
algorithm manages to increase the granularity of the transmitted data. In this sense, the
communication relies on a series of smaller hypercubes, as opposed to a single hypercube
containing the entire, undivided frames form a sequence. This translates into increased
adaptability to the variations of the transmitting channel’s bandwidth.
In the literature there are a wide variety of algorithms for image binarization, the difference between them being the method that identifies the pi xel threshold value. They can be split into two classes: algorithms that use a single... more
In the literature there are a wide variety of algorithms for image binarization, the difference
between them being the method that identifies the pi xel threshold value. They can be split
into two classes: algorithms that use a single threshold for the entire image and tend to
identify a few large objects and algorithms that do the processing in localities and tend to
identify many small items )). Thi s paper aims at defining a method for image thresholding
based on the results of several different algorithms. Knowing in advance the behavior of
specific algorithms on different kinds of images, we can vote between their results. The end
result of the pro posed method is a mosaic of more binarization algorithms, hopefully better
than any individual image.
between them being the method that identifies the pi xel threshold value. They can be split
into two classes: algorithms that use a single threshold for the entire image and tend to
identify a few large objects and algorithms that do the processing in localities and tend to
identify many small items )). Thi s paper aims at defining a method for image thresholding
based on the results of several different algorithms. Knowing in advance the behavior of
specific algorithms on different kinds of images, we can vote between their results. The end
result of the pro posed method is a mosaic of more binarization algorithms, hopefully better
than any individual image.
Although the technology of optical instruments is constantly advancing, the capture of high resolution images is limited by both the shortcoming s of the imaging devices and the law of physics (uncertainty principle applied onto photons... more
Although the technology of optical instruments is constantly advancing, the capture of high
resolution images is limited by both the shortcoming s of the imaging devices and the law of
physics (uncertainty principle applied onto photons or the wave like theory of light). The
current paper presents an algorithm for processing a set of images sharing the same subject
with the purpose of extracting a higher resolution output image of the subject, using partial
information from every one of the low resolutions samples in the input set.
resolution images is limited by both the shortcoming s of the imaging devices and the law of
physics (uncertainty principle applied onto photons or the wave like theory of light). The
current paper presents an algorithm for processing a set of images sharing the same subject
with the purpose of extracting a higher resolution output image of the subject, using partial
information from every one of the low resolutions samples in the input set.
This paper describes an image binarization method that applies localized Otsu thresholding to irregular regions of images, determined through watershed segmentation. Traditional localized binarization techniques work on square regions,... more
This paper describes an image binarization method that applies localized Otsu thresholding
to irregular regions of images, determined through watershed segmentation. Traditional localized
binarization techniques work on square regions, which group together pixels of dierent
origins. This is a problem, as there may not always be available a comparator for objects from
dierent classes. In order to solve this problem, we use a method that rst selects the objects
(like shadow and light regions) and then performs binarization on each individual object in
order to expose its characteristics. For more accurate results, an inter-scale segmentation and
binarization method is proposed.
to irregular regions of images, determined through watershed segmentation. Traditional localized
binarization techniques work on square regions, which group together pixels of dierent
origins. This is a problem, as there may not always be available a comparator for objects from
dierent classes. In order to solve this problem, we use a method that rst selects the objects
(like shadow and light regions) and then performs binarization on each individual object in
order to expose its characteristics. For more accurate results, an inter-scale segmentation and
binarization method is proposed.