CN107103595A - Method, device, storage medium and the equipment of detection image change - Google Patents

Abstract

The invention discloses method, device, storage medium and the equipment of a kind of detection image change, it is related to technical field of image processing, it is possible to increase the degree of accuracy of detection SAR image situation of change.The method of the present invention mainly includes：Obtain the disparity map of two width image to be detected；Frequency band parsing is carried out to the disparity map, low frequency sub-band and high-frequency sub-band is obtained；Linear enhancing processing is carried out to the low frequency sub-band, denoising is carried out to the high-frequency sub-band；Low frequency sub-band after processing and the high-frequency sub-band after processing are merged, the disparity map after being handled；Clustering processing is carried out to the disparity map after the processing, change information and non-changing information between described two width image to be detected is obtained.In the scene detected the present invention is mainly suitable for the situation of change to SAR image.

Description

Method, device, storage medium and equipment for detecting image changes

technical field

The present invention relates to the technical field of image processing, in particular to a method, device, storage medium and equipment for detecting image changes.

Background technique

With the development of science and technology, there are more and more technologies for collecting images. For example, a camera may be used to collect images, and a radar may also be used to collect images. In order to obtain images with higher resolution under weather conditions with extremely low visibility, a synthetic aperture radar (Synthetic Aperture Radar, SAR) has been invented. SAR can be used to observe target objects (such as the surface) all day and all day, collect high-resolution images at different times, and then analyze these images to obtain image change results. However, in the process of SAR image imaging, it will inevitably be affected by multiplicative noise, which makes each SAR image have a large difference from the actual target object, and then greatly reduces the accuracy of detecting image changes.

At present, there are two main methods for detecting image changes: (1) First obtain prior knowledge about the target object (some images less affected by multiplicative noise), then train these prior knowledge to obtain a training model, and finally Based on the training model, detect the change of the image to be detected; (2) first obtain the difference map of the two images to be detected, and then cluster the pixels of the difference map to find out the changed class and the non-changed class.

For method (1), since method (1) is based on relatively accurate prior knowledge to analyze changes in SAR images, even if the SAR image to be detected is greatly affected by multiplicative noise, it will not affect The results of SAR image change detection, but in most cases, prior knowledge cannot be obtained, so the practicability of this method is low. For method (2), since method (2) directly clusters the difference maps of the two images to be detected to obtain the image change results, when the two images are greatly affected by multiplicative noise, there will be The accuracy of the image change detection result is greatly reduced. Therefore, how to improve the accuracy of SAR image change detection results needs to be solved urgently.

Contents of the invention

In view of this, the present invention provides a method, device, storage medium and equipment for detecting image changes, the main purpose of which is to solve the problem of low accuracy in detecting SAR image changes in the prior art.

In order to solve the above problems, the present invention mainly provides the following technical solutions:

In a first aspect, the present invention provides a method for detecting image changes, the method comprising:

Obtain the difference map of the two images to be detected;

performing frequency band analysis on the difference map to obtain low frequency subbands and high frequency subbands;

performing linear enhancement processing on the low-frequency sub-band, and performing denoising processing on the high-frequency sub-band;

Merging the processed low-frequency sub-bands and the processed high-frequency sub-bands to obtain a processed difference map;

Clustering is performed on the processed difference map to obtain change information and non-change information between the two images to be detected.

Optionally, performing frequency band analysis on the difference map to obtain the low-frequency sub-band and the high-frequency sub-band includes:

Performing frequency band analysis on the difference map by non-subsampling Shearlet transform to obtain the low frequency subband and the high frequency subband;

Or, performing frequency band analysis on the difference map by wavelet transform to obtain the low frequency sub-band and the high frequency sub-band;

Alternatively, perform frequency band analysis on the difference map through non-subsampling Contourlet transformation to obtain the low frequency subband and the high frequency subband.

Optionally, the merging the processed low-frequency sub-bands and the processed high-frequency sub-bands to obtain the processed difference map includes:

When the frequency band analysis is performed on the difference map through the non-subsampling Shearlet transform, the processed low-frequency sub-band and the processed high-frequency sub-band are combined through the inverse transform of the non-sub-sampled Shearlet transform to obtain the The difference map after the above processing;

When band analysis is performed on the difference map by wavelet transform, the processed low-frequency sub-band and the processed high-frequency sub-band are combined by inverse wavelet transform to obtain the processed difference map ;

When the frequency band analysis is performed on the difference map through the non-subsampled Contourlet transform, the processed low-frequency sub-band and the processed high-frequency sub-band are combined through the inverse transform of the non-sub-sampled Contourlet transform to obtain the The difference map after the above processing.

Optionally, the performing denoising processing on the high frequency sub-band includes:

The noise in the high frequency sub-band is suppressed by using an adaptive threshold method.

Optionally, the clustering the processed difference map includes:

The processed difference map is clustered using a fuzzy local information C-means clustering algorithm.

Optionally, the obtaining the difference map of the two images to be detected includes:

performing smoothing filtering and denoising processing on the two images to be detected;

The difference map is obtained by comparing the two images to be detected after denoising by using a normalized neighborhood ratio method.

In a second aspect, the present invention provides a device for detecting image changes, the device comprising:

An acquisition unit, configured to acquire a difference map of two images to be detected;

An analysis unit, configured to perform frequency band analysis on the difference map acquired by the acquisition unit, to obtain low-frequency sub-bands and high-frequency sub-bands;

a processing unit, configured to perform linear enhancement processing on the low-frequency sub-band obtained by the analysis unit, and perform denoising processing on the high-frequency sub-band obtained by the analysis unit;

a merging unit, configured to combine the low-frequency sub-bands processed by the processing unit and the processed high-frequency sub-bands to obtain a processed difference map;

A clustering unit is configured to perform clustering processing on the processed difference map obtained by the merging unit to obtain change information and non-change information between the two images to be detected.

Optionally, the parsing unit includes:

A first parsing module, configured to perform frequency band parsing on the difference map through non-subsampling Shearlet transform, to obtain the low frequency subband and the high frequency subband;

The second parsing module is configured to perform frequency band parsing on the difference map by wavelet transform to obtain the low frequency subband and the high frequency subband;

The third parsing module is configured to perform frequency band parsing on the difference map through non-subsampling Contourlet transform to obtain the low frequency sub-band and the high frequency sub-band.

Optionally, the merging unit includes:

The first merging module is used to perform frequency band analysis on the difference map through non-subsampling Shearlet transform, and perform the inverse transformation of non-subsampling Shearlet transform on the processed low-frequency sub-band and the processed high-frequency sub-band The sub-bands are merged to obtain the processed difference map;

The second merging module is configured to combine the processed low-frequency sub-bands and the processed high-frequency sub-bands through wavelet inverse transform to obtain The difference map after the processing;

The third merging module is used to perform frequency band analysis on the difference map through non-subsampling Contourlet transformation, and perform the inverse transformation of non-subsampling Contourlet transformation on the processed low-frequency sub-band and the processed high-frequency sub-band The subbands are merged to obtain the processed difference map.

Optionally, the processing unit is configured to use an adaptive threshold method to suppress noise in the high frequency sub-band.

Optionally, the clustering unit is configured to cluster the processed difference map by using a fuzzy local information C-means clustering algorithm.

Optionally, the acquisition unit includes:

A denoising module, configured to perform smoothing filtering and denoising processing on the two images to be detected;

The comparison module is configured to use the normalized neighborhood ratio method to compare the two images to be detected after denoising by the denoising module to obtain the difference map.

In a third aspect, the present invention provides a storage medium, the storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method for detecting image changes as described in the first aspect.

In a fourth aspect, the present invention provides a device, the device comprising:

a processor adapted to implement the instructions; and

a storage medium adapted to store a plurality of instructions;

The instructions are suitable for being loaded by the processor and executing the method for detecting image changes as described in the first aspect.

By means of the above technical solution, the technical solution provided by the present invention has at least the following advantages:

Compared with the prior art, the method, device, storage medium and equipment provided by the present invention for detecting image changes not only do not need to obtain prior knowledge, but also do not directly perform the difference map after obtaining the difference map of two images. Clustering processing in order to obtain image changes, but first perform frequency band analysis on the difference image to obtain the low-frequency sub-bands that contain the most effective information, the most important and include the least noise, and the high-frequency sub-bands that include the least effective information and the most noise band, and then linearly enhance the low-frequency sub-band to enhance the display strength of effective information, and perform denoising processing on the high-frequency sub-band to remove a large amount of noise in the high-frequency sub-band, and then the processed low-frequency sub-band Merge with high-frequency sub-bands to obtain a difference map with effective information enhancement and noise information reduction, and finally cluster the processed difference map, thereby improving the accuracy of detecting image changes.

The above description is only an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the description, and in order to make the above and other purposes, features and advantages of the present invention more obvious and understandable , the specific embodiments of the present invention are enumerated below.

Description of drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiment. The drawings are only for the purpose of illustrating a preferred embodiment and are not to be considered as limiting the invention. Also throughout the drawings, the same reference symbols are used to designate the same parts. In the attached picture:

FIG. 1 shows a flow chart of a method for detecting image changes provided by an embodiment of the present invention;

FIG. 2 shows an example diagram of a method for detecting image changes provided by an embodiment of the present invention;

FIG. 3 shows a block diagram of a device for detecting image changes provided by an embodiment of the present invention;

Fig. 4 shows a block diagram of a device for detecting image changes provided by an embodiment of the present invention.

detailed description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided for more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

The embodiment of the present invention provides a method for detecting image changes, as shown in Figure 1, the method mainly includes:

101. Acquire a difference map of two images to be detected.

After the images of the target object (that is, the object to be photographed) at different moments are collected by radar and other equipment, the change between any two images can be detected. Specifically, since the target object often does not change within a period of time, when detecting image changes, two images with a certain time interval are often selected for detection.

After obtaining the two images to be detected, the two images can be directly compared to obtain the difference map between the two, but in order to make the obtained difference map more accurate, the two images can be denoised by smoothing filter first. Preliminary denoising is performed on each image to remove a large amount of additive noise, and then the denoised images are compared to obtain a difference map.

When obtaining the difference map of the two images, the method adopted may be a difference method, a ratio method, or other methods. Among them, the ratio method includes the normalized field ratio method.

102. Perform frequency band analysis on the difference map to obtain low frequency subbands and high frequency subbands.

An image is often composed of low-frequency sub-bands and high-frequency sub-bands, and the low-frequency sub-bands often contain a large amount of important and effective information, and the low-frequency sub-bands contain less noise, and the high-frequency sub-bands contain There is less effective information (mainly image edge information) and more noise. Therefore, after obtaining the difference map, the difference map can be analyzed by frequency band first, and the low-frequency sub-band and high-frequency sub-band can be obtained from it, and then the low-frequency sub-band and high-frequency sub-band can be obtained respectively. The characteristics of the high-frequency sub-band are analyzed and processed, that is, step 103 is executed.

Among them, the specific algorithms for performing frequency band analysis on the difference map to obtain low-frequency sub-bands and high-frequency sub-bands include non-subsampled Shearlet transform (NSST transform), wavelet transform, and non-subsampled Contourlet transform (NSCT transform). That is to say, the specific implementation of this step may be: performing frequency band analysis on the difference map through NSST transformation, to obtain the low-frequency sub-band and the high-frequency sub-band;

Or, performing frequency band analysis on the difference map by wavelet transform to obtain the low frequency sub-band and the high frequency sub-band;

Alternatively, band analysis is performed on the difference map by NSCT to obtain the low-frequency sub-band and the high-frequency sub-band.

What needs to be added is that for the difference map of the SAR image, the frequency band analysis efficiency of the difference map using the NSST transform is higher than that of other methods. Therefore, the effect of using the NSST transform in this step is better.

103. Perform linear enhancement processing on the low frequency subband, and perform denoising processing on the high frequency subband.

As mentioned in the above step 102, the low-frequency sub-band often contains a large amount of important and effective information, and the low-frequency sub-band contains less noise, and the high-frequency sub-band contains less effective information and more noise, so for To make the display intensity of the effective information stronger and reduce the noise in the image as much as possible, the low-frequency sub-band can be linearly enhanced, and the high-frequency sub-band can be denoised.

Wherein, the intensity of the linear enhancement performed on the low frequency sub-band can be determined according to the characteristics of the image itself, for example, if the display of the image in the low frequency sub-band to be processed is weak, the intensity of the linear enhancement can be stronger. The noise denoising the high-frequency sub-band is mainly multiplicative noise, and there are many denoising methods, such as adaptive threshold method, median filter, mean filter and so on.

104. Merge the processed low-frequency subbands and the processed high-frequency subbands to obtain a processed difference map.

After performing linear enhancement processing on the low-frequency sub-band and denoising processing on the high-frequency sub-band, the processed low-frequency sub-band and the processed high-frequency sub-band can be combined to obtain effective information enhancement and noise information reduction. After the difference map, image change detection is performed.

When the low-frequency sub-bands and high-frequency sub-bands are recombined into a difference map, the algorithm adopted is the inverse operation of the algorithm used for decomposing the difference map into low-frequency sub-bands and high-frequency sub-bands. That is to say, when band analysis is performed on the difference map through NSST transform, the processed low-frequency sub-band and the processed high-frequency sub-band are combined through an inverse transform of NSST transform to obtain the processed After the difference map; when the frequency band analysis is performed on the difference map by wavelet transform, the processed low-frequency sub-band and the processed high-frequency sub-band are combined through the inverse transform of wavelet transform to obtain the described The processed difference map; when the frequency band analysis is performed on the difference map by NSCT transform, the processed low-frequency sub-band and the processed high-frequency sub-band are combined by the inverse transform of NSCT transform to obtain the obtained The difference map after the above processing.

105. Perform clustering processing on the processed difference map to obtain change information and non-change information between the two images to be detected.

After obtaining the difference map with effective information enhancement and noise information reduction, the clustering algorithm can be used to cluster the pixels in the difference map to obtain the change class and non-change class, that is, the change information and non-changing information. Wherein, the clustering algorithm includes but not limited to the following algorithms: fuzzy local information C-means clustering (FLICM) algorithm, fuzzy C-means clustering (FCM) algorithm and kernel function-based fuzzy c-means (KFCM) algorithm.

It should be added that the image detected in the embodiment of the present invention may be a SAR image or other images, and its specific type is not limited here.

Compared with the prior art, the method for detecting image changes provided by the embodiment of the present invention not only does not need to obtain prior knowledge, but also does not directly perform clustering processing on the difference map after obtaining the difference map of two images, so that To obtain the image changes, first perform frequency band analysis on the difference image to obtain the low-frequency sub-bands that include the most effective information, the most important and the least noise, and the high-frequency sub-bands that include the least effective information and the most noise, and then analyze the low-frequency The sub-band is linearly enhanced to enhance the display intensity of effective information, and the high-frequency sub-band is de-noised to remove a large amount of noise in the high-frequency sub-band, and then the processed low-frequency sub-band and high-frequency sub-band Merging is performed to obtain a difference map with effective information enhancement and noise information reduction, and finally the processed difference map is clustered, thereby improving the accuracy of detecting image changes.

Taking the image to be detected as a SAR image, and the algorithms used include the normalized neighborhood ratio method, NSST transform, adaptive threshold method, and FLICM algorithm as examples, the above methods for detecting image changes are described, as shown in Figure 2. The method mainly includes:

201. Perform smoothing filtering and denoising processing on SAR image A to obtain SAR image a; perform smoothing filtering and denoising processing on SAR image B to obtain SAR image b;

202. Using the normalized neighborhood ratio method to compare the SAR image a and the SAR image b to obtain a difference map M;

203. Perform frequency band analysis on the difference map M through NSST transformation, and obtain low-frequency sub-band c and high-frequency sub-band d;

204. Perform linear enhancement processing on the low-frequency sub-band c to obtain the low-frequency sub-band c1; suppress noise in the high-frequency sub-band d by using an adaptive threshold method to obtain the high-frequency sub-band d1;

205. Merge the low-frequency subband c1 and the high-frequency subband d1 through the inverse transformation of the NSST transformation to obtain a difference map N;

206. Use the FLICM algorithm to perform clustering processing on the difference map N to obtain a change class n1 and a non-change class n2.

Among them, the change category n1 and the non-change category n2 are the changes between SAR image A and SAR image B.

Further, according to the above-mentioned method embodiment, another embodiment of the present invention also provides a device for detecting image changes, as shown in FIG. 3 , the device mainly includes: an acquisition unit 31, an analysis unit 32, and a processing unit 33 , a merging unit 34 and a clustering unit 35 . in,

An acquisition unit 31, configured to acquire a difference map of two images to be detected;

An analysis unit 32, configured to perform frequency band analysis on the difference map acquired by the acquisition unit 31, to obtain low-frequency sub-bands and high-frequency sub-bands;

A processing unit 33, configured to perform linear enhancement processing on the low-frequency sub-band obtained by the analysis unit 32, and perform denoising processing on the high-frequency sub-band obtained by the analysis unit 32;

A merging unit 34, configured to combine the low-frequency sub-bands processed by the processing unit 33 and the processed high-frequency sub-bands to obtain a processed difference map;

The clustering unit 35 is configured to perform clustering processing on the processed difference map obtained by the merging unit 34 to obtain change information and non-change information between the two images to be detected.

Optionally, as shown in Figure 4, the parsing unit 32 includes:

The first parsing module 321 is configured to perform frequency band parsing on the difference map through non-subsampling Shearlet transform, to obtain the low frequency subband and the high frequency subband;

The second parsing module 322 is configured to perform frequency band parsing on the difference map by wavelet transform to obtain the low frequency subband and the high frequency subband;

The third parsing module 323 is configured to perform frequency band parsing on the difference map through non-subsampling Contourlet transform, to obtain the low frequency subband and the high frequency subband.

Optionally, as shown in FIG. 4, the merging unit 34 includes:

The first merging module 341 is configured to perform frequency band analysis on the difference map through non-subsampling Shearlet transform, perform inverse transformation of non-subsampling Shearlet transform on the processed low-frequency sub-band and the processed high-frequency sub-band The frequency sub-bands are combined to obtain the processed difference map;

The second merging module 342 is configured to merge the processed low-frequency sub-bands and the processed high-frequency sub-bands through inverse wavelet transform when performing frequency band analysis on the difference map through wavelet transform, obtaining the processed difference map;

The third merging module 343 is configured to perform frequency band analysis on the difference map through non-subsampling Contourlet transform, perform inverse transformation of non-subsampling Contourlet transform on the processed low-frequency sub-band and the processed high-frequency sub-band The frequency subbands are combined to obtain the processed difference map.

Optionally, the processing unit 33 is configured to suppress noise in the high-frequency sub-band by using an adaptive threshold method.

Optionally, the clustering unit 35 is configured to cluster the processed difference map by using a fuzzy local information C-means clustering algorithm.

Optionally, as shown in FIG. 4, the acquisition unit 31 includes:

A denoising module 311, configured to perform smoothing filtering and denoising processing on the two images to be detected;

The comparison module 312 is configured to use the normalized neighborhood ratio method to compare the two images to be detected after denoising by the denoising module 311 to obtain the difference map.

Compared with the prior art, the device for detecting image changes provided by the embodiment of the present invention not only does not need to obtain prior knowledge, but also does not directly perform clustering processing on the difference map after obtaining the difference map of two images, so as to To obtain the image changes, first perform frequency band analysis on the difference image to obtain the low-frequency sub-bands that include the most effective information, the most important and the least noise, and the high-frequency sub-bands that include the least effective information and the most noise, and then analyze the low-frequency The sub-band is linearly enhanced to enhance the display intensity of effective information, and the high-frequency sub-band is de-noised to remove a large amount of noise in the high-frequency sub-band, and then the processed low-frequency sub-band and high-frequency sub-band Merging is performed to obtain a difference map with effective information enhancement and noise information reduction, and finally the processed difference map is clustered, thereby improving the accuracy of detecting image changes.

It should be noted that this device embodiment corresponds to the foregoing method embodiment. For the convenience of reading, this device embodiment will not repeat the details of the foregoing method embodiment one by one, but it should be clear that the device in this embodiment can Correspondingly implement all the contents in the foregoing method embodiments.

The device for detecting image changes includes a processor and a memory. The acquisition unit, analysis unit, processing unit, merging unit, and clustering unit are all stored in the memory as program units, and the processor executes the above-mentioned program stored in the memory. unit to achieve the corresponding function.

The processor includes a kernel, and the kernel fetches corresponding program units from the memory. One or more kernels can be set, and the accuracy of detecting changes in SAR images can be improved by adjusting kernel parameters.

Memory may include non-permanent memory in computer-readable media, random access memory (RAM) and/or non-volatile memory, such as read-only memory (ROM) or flash memory (flashRAM), and memory includes at least one memory chip.

An embodiment of the present invention provides a storage medium, the storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the above-mentioned method for detecting image changes.

Compared with the prior art, the storage medium for detecting image changes provided by the embodiment of the present invention not only does not need to acquire prior knowledge, but also does not directly perform clustering processing on the difference map after obtaining the difference map of two images. In order to obtain the image changes, the difference image is firstly analyzed by frequency bands to obtain the low-frequency sub-bands that contain the most effective information, the most important and the least noise, and the high-frequency sub-bands that contain the least effective information and the most noise, and then The low-frequency sub-band is linearly enhanced to enhance the display intensity of effective information, and the high-frequency sub-band is de-noised to remove a large amount of noise in the high-frequency sub-band, and then the processed low-frequency sub-band and high-frequency sub-band Bands are combined to obtain a difference map with effective information enhancement and noise information reduction, and finally the processed difference map is clustered, thereby improving the accuracy of detecting image changes.

An embodiment of the present invention provides a processor, and the processor is configured to run a program, wherein the above-mentioned method for detecting image changes is executed when the program is running.

An embodiment of the present invention provides a device, and the device includes a processor and a storage medium. Wherein, the processor is adapted to implement each instruction; and the storage medium is adapted to store multiple instructions; the instructions are adapted to be loaded and executed by the processor:

Obtain the difference map of the two images to be detected;

performing frequency band analysis on the difference map to obtain low frequency subbands and high frequency subbands;

performing linear enhancement processing on the low-frequency sub-band, and performing denoising processing on the high-frequency sub-band;

Merging the processed low-frequency sub-bands and the processed high-frequency sub-bands to obtain a processed difference map;

Clustering is performed on the processed difference map to obtain change information and non-change information between the two images to be detected.

Optionally, performing frequency band analysis on the difference map to obtain the low-frequency sub-band and the high-frequency sub-band includes:

Performing frequency band analysis on the difference map by non-subsampling Shearlet transform to obtain the low frequency subband and the high frequency subband;

Or, performing frequency band analysis on the difference map by wavelet transform to obtain the low frequency sub-band and the high frequency sub-band;

Alternatively, perform frequency band analysis on the difference map through non-subsampling Contourlet transformation to obtain the low frequency subband and the high frequency subband.

Optionally, the merging the processed low-frequency sub-bands and the processed high-frequency sub-bands to obtain the processed difference map includes:

When the frequency band analysis is performed on the difference map through the non-subsampling Shearlet transform, the processed low-frequency sub-band and the processed high-frequency sub-band are combined through the inverse transform of the non-sub-sampled Shearlet transform to obtain the The difference map after the above processing;

When band analysis is performed on the difference map by wavelet transform, the processed low-frequency sub-band and the processed high-frequency sub-band are combined by inverse wavelet transform to obtain the processed difference map ;

When the frequency band analysis is performed on the difference map through the non-subsampled Contourlet transform, the processed low-frequency sub-band and the processed high-frequency sub-band are combined through the inverse transform of the non-sub-sampled Contourlet transform to obtain the The difference map after the above processing.

Optionally, the performing denoising processing on the high frequency sub-band includes:

The noise in the high frequency sub-band is suppressed by using an adaptive threshold method.

Optionally, the clustering the processed difference map includes:

The processed difference map is clustered using a fuzzy local information C-means clustering algorithm.

Optionally, the obtaining the difference map of the two images to be detected includes:

performing smoothing filtering and denoising processing on the two images to be detected;

The difference map is obtained by comparing the two images to be detected after denoising by using a normalized neighborhood ratio method.

It should be noted that the devices in this article may be servers, PCs, PADs, mobile phones, etc.

Compared with the prior art, the device for detecting image changes provided by the embodiment of the present invention not only does not need to obtain prior knowledge, but also does not directly perform clustering processing on the difference map after obtaining the difference map of two images, so that To obtain the image changes, first perform frequency band analysis on the difference image to obtain the low-frequency sub-bands that include the most effective information, the most important and the least noise, and the high-frequency sub-bands that include the least effective information and the most noise, and then analyze the low-frequency The sub-band is linearly enhanced to enhance the display intensity of effective information, and the high-frequency sub-band is de-noised to remove a large amount of noise in the high-frequency sub-band, and then the processed low-frequency sub-band and high-frequency sub-band Merging is performed to obtain a difference map with effective information enhancement and noise information reduction, and finally the processed difference map is clustered, thereby improving the accuracy of detecting image changes.

The present application also provides a computer program product, which, when executed on a data processing device, is adapted to execute a program code initialized with the following method steps:

Obtain the difference map of the two images to be detected;

performing frequency band analysis on the difference map to obtain low frequency subbands and high frequency subbands;

performing linear enhancement processing on the low-frequency sub-band, and performing denoising processing on the high-frequency sub-band;

Merging the processed low-frequency sub-bands and the processed high-frequency sub-bands to obtain a processed difference map;

Clustering is performed on the processed difference map to obtain change information and non-change information between the two images to be detected.

Those skilled in the art should understand that the embodiments of the present application may be provided as methods, systems, or computer program products. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present application is described with reference to flowcharts and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for realizing the functions specified in one or more steps of the flow chart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions The device implements the functions specified in one or more frames of the flow chart and/or one or more blocks of the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby The instructions provide steps for implementing the functions specified in the flow diagram or flow diagrams and/or the block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

Memory may include non-permanent storage in computer readable media, in the form of random access memory (RAM) and/or nonvolatile memory, such as read only memory (ROM) or flash RAM. The memory is an example of a computer readable medium.

Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can be implemented by any method or technology for storage of information. Information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Flash memory or other memory technology, Compact Disc Read-Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, Magnetic tape cartridge, tape magnetic disk storage or other magnetic storage device or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media excludes transitory computer-readable media, such as modulated data signals and carrier waves.

It should also be noted that the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes Other elements not expressly listed, or elements inherent in the process, method, commodity, or apparatus are also included. Without further limitations, an element defined by the phrase "comprising a piece of ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

Those skilled in the art should understand that the embodiments of the present application may be provided as methods, systems or computer program products. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The above are only examples of the present application, and are not intended to limit the present application. For those skilled in the art, various modifications and changes may occur in this application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims (10)

1. A method of detecting image changes, the method comprising:

acquiring difference graphs of two images to be detected;

performing frequency band analysis on the difference graph to obtain a low-frequency sub-band and a high-frequency sub-band;

carrying out linear enhancement processing on the low-frequency sub-band, and carrying out denoising processing on the high-frequency sub-band;

merging the processed low-frequency sub-band and the processed high-frequency sub-band to obtain a processed difference map;

and clustering the processed difference graph to obtain the change information and the non-change information between the two images to be detected.

2. The method of claim 1, wherein the band-parsing the difference map to obtain low-frequency subbands and high-frequency subbands comprises:

performing band analysis on the difference map through non-downsampling Shearlet transformation to obtain the low-frequency sub-band and the high-frequency sub-band;

or performing band analysis on the difference map through wavelet transformation to obtain the low-frequency sub-band and the high-frequency sub-band;

or performing band analysis on the difference map through non-subsampled Contourlet transformation to obtain the low-frequency sub-band and the high-frequency sub-band.

3. The method of claim 2, wherein the combining the processed low frequency subbands and the processed high frequency subbands to obtain the processed difference map comprises:

when the frequency band of the difference map is analyzed through the non-downsampling Shearlet transform, combining the processed low-frequency sub-band and the processed high-frequency sub-band through the inverse transform of the non-downsampling Shearlet transform to obtain the processed difference map;

when the difference map is subjected to band analysis through wavelet transformation, combining the processed low-frequency sub-band and the processed high-frequency sub-band through inverse transformation of wavelet transformation to obtain the processed difference map;

and when the frequency band of the difference map is analyzed through the non-downsampling Contourlet transformation, combining the processed low-frequency sub-band and the processed high-frequency sub-band through the inverse transformation of the non-downsampling Contourlet transformation to obtain the processed difference map.

4. The method of claim 1, wherein denoising the high frequency subband comprises:

and suppressing noise in the high-frequency sub-band by using an adaptive threshold method.

5. The method according to any one of claims 1 to 4, wherein the clustering the processed disparity map comprises:

and clustering the processed difference graph by using a fuzzy local information C mean clustering algorithm.

6. The method according to any one of claims 1 to 4, wherein said obtaining a disparity map of two images to be detected comprises:

carrying out smooth filtering denoising processing on the two images to be detected;

and comparing the two images to be detected after denoising by utilizing a normalized neighborhood ratio method to obtain the difference image.

7. An apparatus for detecting image changes, the apparatus comprising:

the acquiring unit is used for acquiring difference images of two images to be detected;

the analysis unit is used for carrying out frequency band analysis on the difference map acquired by the acquisition unit to acquire a low-frequency sub-band and a high-frequency sub-band;

the processing unit is used for carrying out linear enhancement processing on the low-frequency sub-band obtained by the analysis unit and carrying out denoising processing on the high-frequency sub-band obtained by the analysis unit;

a merging unit, configured to merge the low-frequency subband processed by the processing unit and the processed high-frequency subband to obtain a processed difference map;

and the clustering unit is used for clustering the processed difference graph obtained by the merging unit to obtain the change information and the non-change information between the two images to be detected.

8. The apparatus of claim 7, wherein the parsing unit comprises:

the first analysis module is used for carrying out frequency band analysis on the difference map through non-downsampling Shearlet transformation to obtain the low-frequency sub-band and the high-frequency sub-band;

the second analysis module is used for carrying out frequency band analysis on the difference map through wavelet transformation to obtain the low-frequency sub-band and the high-frequency sub-band;

and the third analysis module is used for carrying out frequency band analysis on the difference graph through non-subsampled Contourlet conversion to obtain the low-frequency sub-band and the high-frequency sub-band.

9. A storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform a method of detecting image changes according to any one of claims 1 to 6.

10. An apparatus, characterized in that the apparatus comprises:

a processor adapted to implement instructions; and

a storage medium adapted to store a plurality of instructions;

the instructions are adapted to be loaded by the processor and to perform the method of detecting image changes according to any of claims 1 to 6.