CN112819796A - Tobacco shred foreign matter identification method and equipment - Google Patents
Tobacco shred foreign matter identification method and equipment Download PDFInfo
- Publication number
- CN112819796A CN112819796A CN202110160109.1A CN202110160109A CN112819796A CN 112819796 A CN112819796 A CN 112819796A CN 202110160109 A CN202110160109 A CN 202110160109A CN 112819796 A CN112819796 A CN 112819796A
- Authority
- CN
- China
- Prior art keywords
- tobacco shred
- foreign matter
- image
- tobacco
- sample data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 235000002637 Nicotiana tabacum Nutrition 0.000 title claims abstract description 120
- 238000000034 method Methods 0.000 title claims abstract description 45
- 244000061176 Nicotiana tabacum Species 0.000 title 1
- 241000208125 Nicotiana Species 0.000 claims abstract description 119
- 238000001514 detection method Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 230000004927 fusion Effects 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 10
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000002372 labelling Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 5
- 230000002776 aggregation Effects 0.000 claims description 4
- 238000004220 aggregation Methods 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000012946 outsourcing Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000003860 storage Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 229920003023 plastic Polymers 0.000 description 2
- 239000004033 plastic Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 235000019580 granularity Nutrition 0.000 description 1
- 238000013140 knowledge distillation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Analysis (AREA)
Abstract
The application relates to a method and equipment for identifying foreign matters in tobacco shreds, comprising the following steps: and acquiring the tobacco shred image acquired by the image acquisition device, and preprocessing the tobacco shred image to enable the tobacco shred image to be more easily identified. Obtaining a recognition result of the tobacco shred image based on a pre-trained tobacco shred foreign matter recognition model according to the preprocessed tobacco shred image; the recognition result at least comprises: whether foreign matters exist in the tobacco shred image or not, the type of the foreign matters and the coordinate positions of the foreign matters. Can effectively utilize a large amount of picture data to train pipe tobacco foreign matter recognition model earlier in this application, pipe tobacco foreign matter recognition model possesses high detection precision and generalization ability, and whether there is the foreign matter in the detection pipe tobacco image that can be accurate, foreign matter classification and foreign matter coordinate position.
Description
Technical Field
The application relates to the technical field of image recognition, in particular to a method and equipment for recognizing foreign matters in tobacco shreds.
Background
In the tobacco shred production process, foreign matters such as paper scraps, plastics and the like and foreign matters such as stems, leaves, wet lumps and the like can be mixed, for the foreign matters, an effective online monitoring method does not exist in the prior art, algorithms such as filtering and threshold segmentation are generally adopted, the algorithms depend too much on manual debugging algorithm parameters, a good effect can be achieved for a specific scene, but a large amount of picture data cannot be utilized, relatively good generalization capability does not exist, and great disturbance can possibly occur when the environment or the tobacco shreds are slightly changed, so that the algorithms cannot work normally.
Disclosure of Invention
In order to overcome the problem that foreign matters mixed in the tobacco shred production process cannot be effectively identified in the related technology at least to a certain extent, the application provides a tobacco shred foreign matter identification method and equipment.
The scheme of the application is as follows:
according to a first aspect of the embodiments of the present application, there is provided a method for identifying cut tobacco foreign matters, including:
acquiring a tobacco shred image acquired by an image acquisition device;
preprocessing the tobacco shred image;
obtaining a recognition result of the tobacco shred image based on a pre-trained tobacco shred foreign matter recognition model according to the preprocessed tobacco shred image; the recognition result at least comprises: whether foreign matters exist in the tobacco shred images or not, the types of the foreign matters and the coordinate positions of the foreign matters.
Preferably, in an implementation manner of the present application, the method further includes:
and acquiring a tobacco shred image doped with foreign matters as sample data, and training the tobacco shred foreign matter identification model.
Preferably, in an implementation manner of the present application, the training of the tobacco shred foreign object identification model specifically includes:
marking the sample data based on a marking tool;
establishing the tobacco shred foreign matter identification model by adopting a deep neural network target detection framework;
inputting the marked sample data into the tobacco shred foreign matter identification model;
extracting the characteristics of the marked sample data based on the tobacco shred foreign matter identification model;
performing feature fusion on the extracted features based on the tobacco shred foreign matter identification model, and outputting a plurality of feature maps with different scales;
obtaining a plurality of prediction results of different scales for the feature maps of different scales through a prediction module; the prediction result at least comprises: and whether the sample data contains foreign matters, the types of the foreign matters and the coordinate positions of the foreign matters or not.
Preferably, in an implementation manner of the present application, the labeling the sample data based on the labeling tool specifically includes:
identifying foreign objects in the sample data by using a minimum outsourcing rectangular frame;
establishing a coordinate system by taking the sample data as a reference, and acquiring the coordinate of the central point of the foreign matter in the coordinate system;
acquiring the length and the width of the minimum external rectangular frame of the foreign matters;
identifying a type of the foreign matter.
Preferably, in an implementation manner of the present application, before inputting the labeled sample data into the tobacco shred foreign object identification model, the method further includes:
and performing image enhancement on the labeled sample data based on image occlusion and multi-image combination modes.
Preferably, in an implementation manner of the present application, before inputting the labeled sample data into the tobacco shred foreign object identification model, the method further includes:
and carrying out grid division on the labeled sample data.
Preferably, in an implementation manner of the present application, the performing feature extraction on the labeled sample data based on the tobacco shred foreign object identification model specifically includes:
and dividing the feature mapping of the deep convolutional neural network basic layer into two parts, merging the divided two parts of feature mapping through a cross-stage hierarchical structure, and repeatedly executing for multiple times to obtain a feature map.
Preferably, in an implementation manner of the present application, the performing feature fusion on the extracted features based on the tobacco shred foreign object identification model, and outputting a plurality of feature maps with different scales specifically includes:
and carrying out parameter aggregation on different detection layers from different trunk layers of the deep convolutional neural network, improving the capability of feature extraction, and outputting a plurality of feature maps with different scales.
Preferably, in an implementation manner of the present application, the preprocessing the cut tobacco image specifically includes:
segmenting the tobacco shred image with high resolution;
and carrying out grid division on the cut tobacco image after segmentation.
According to a second aspect of the embodiments of the present application, there is provided a tobacco shred foreign matter identification apparatus, including:
a processor and a memory;
the processor and the memory are connected through a communication bus:
the processor is used for calling and executing the program stored in the memory;
the memory is used for storing a program, and the program is at least used for executing the cut tobacco foreign matter identification method.
The technical scheme provided by the application can comprise the following beneficial effects: according to the tobacco shred foreign matter identification method, the tobacco shred image acquired by the image acquisition device is acquired, and the tobacco shred image is preprocessed, so that the tobacco shred image is easier to identify. Obtaining a recognition result of the tobacco shred image based on a pre-trained tobacco shred foreign matter recognition model according to the preprocessed tobacco shred image; the recognition result at least comprises: whether foreign matters exist in the tobacco shred image or not, the type of the foreign matters and the coordinate positions of the foreign matters. Can effectively utilize a large amount of picture data to train pipe tobacco foreign matter recognition model earlier in this application, pipe tobacco foreign matter recognition model possesses high detection precision and generalization ability, and whether there is the foreign matter in the detection pipe tobacco image that can be accurate, foreign matter classification and foreign matter coordinate position.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
Fig. 1 is a schematic flow chart of a method for identifying foreign matters in cut tobacco according to an embodiment of the present application;
fig. 2 is a schematic flow chart of training a tobacco shred foreign matter identification model in a tobacco shred foreign matter identification method according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a cut tobacco foreign matter identification device according to an embodiment of the present application.
Reference numerals: a processor-31; a memory-32.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
A method for identifying foreign matters in tobacco shreds is disclosed, which comprises the following steps:
s11: acquiring a tobacco shred image acquired by an image acquisition device;
in the image acquisition stage, the image acquisition device can be placed on a tobacco shred conveying belt to acquire tobacco shred images at regular time, and after the image acquisition device acquires the images, the acquired tobacco shred images are acquired from the image acquisition device.
Preferably, the image acquisition device adopts a high-resolution and high-speed camera to photograph the cut tobacco on the cut tobacco conveyor belt.
S12: preprocessing the tobacco shred image;
preprocessing the cut tobacco image, which specifically comprises the following steps:
segmenting the high-resolution tobacco shred image; in order to reduce the computational complexity, the high-resolution picture needs to be divided into suitable sizes, usually, multiples of 32, such as 640x640, are taken, so that the number of parameters of the tobacco shred foreign matter identification model is controlled to a reasonable range, and an image processing workstation can process the parameters.
And carrying out mesh division on the cut tobacco image after segmentation. In order to further reduce the computational complexity, the segmented single image is also divided into grids, such as 7 × 7 grids, and then the cut tobacco images divided into grids are identified by using a cut tobacco foreign matter identification model.
S13: obtaining a recognition result of the tobacco shred image based on a pre-trained tobacco shred foreign matter recognition model according to the preprocessed tobacco shred image; the recognition result at least comprises: whether foreign matters exist in the tobacco shred image or not, the type of the foreign matters and the coordinate positions of the foreign matters.
The tobacco shred foreign matter identification method in the embodiment can effectively utilize a large amount of picture data to train the tobacco shred foreign matter identification model, the tobacco shred foreign matter identification model has extremely high detection precision and generalization capability, and whether foreign matters exist in tobacco shred images or not, foreign matter categories and foreign matter coordinate positions can be accurately detected.
The method for identifying foreign matters in tobacco shreds in some embodiments further comprises:
and acquiring a tobacco shred image doped with foreign matters as sample data, and training a tobacco shred foreign matter identification model.
Referring to fig. 2, the method specifically includes:
s21: marking the sample data based on a marking tool;
the method specifically comprises the following steps:
identifying foreign objects in the sample data using a minimum outsourcing rectangular frame; the minimum outer rectangular frame is a minimum outer rectangle which surrounds the foreign matters and is parallel to the x and y axes;
and establishing a coordinate system by taking the sample data as a reference, and acquiring the coordinate of the central point of the foreign matter in the coordinate system.
And acquiring the length and the width of the minimum outsourcing rectangular frame of the foreign matters so as to acquire the size of the foreign matters.
Identifying the type of foreign object, which generally includes: paper scraps, plastics and other foreign matters, and stem slivers, leaves, wet balls and other foreign matters.
S22: establishing a tobacco shred foreign matter identification model by adopting a deep neural network target detection framework;
the tobacco shred foreign matter identification model is mainly divided into three parts: the device comprises a feature extraction module, a feature fusion module and a prediction module.
S23: inputting the marked sample data into a tobacco shred foreign matter identification model;
before inputting the marked sample data into the tobacco shred foreign matter identification model, the method further comprises the following steps:
and performing image enhancement on the labeled sample data based on image occlusion and multi-image combination modes. Before the image is input into the model, the image is enhanced, the number and diversity of samples are increased, and the target detection capability is improved. In this embodiment, preferably, image enhancement is performed on the labeled sample data by using image occlusion, multi-graph combination, and the like, and particularly, a method for enhancing a Mosaic multi-graph combination can be used, so that the detection capability of a small target can be effectively improved.
And carrying out grid division on the labeled sample data.
To further reduce computational complexity, the single sample image is gridded, e.g., into a 7x7 grid.
S24: extracting the characteristics of the marked sample data based on the tobacco shred foreign matter identification model;
and a feature extraction module based on the tobacco shred foreign matter identification model is used for extracting features of the marked sample data, and the module is a deep convolution neural network which is used for aggregating and forming image features on different image fine granularities.
Specifically, a CSP Net cross-stage local network is adopted, and rich information features are extracted from labeled sample data. The CSP module is adopted to divide the feature mapping of the deep convolutional neural network basic layer into two parts, and then the two parts are combined through a cross-stage hierarchical structure, so that the accuracy can be ensured while the calculated amount is reduced. And after the CSP module passes through a plurality of CSP modules, repeatedly executing the CSP modules for a plurality of times to obtain a characteristic diagram.
S25: performing feature fusion on the extracted features based on the tobacco shred foreign matter recognition model, and outputting a plurality of feature maps with different scales;
and the feature fusion module based on the tobacco shred foreign matter recognition model performs feature fusion on the extracted features.
In order to better extract the fused features, a feature fusion module is added between the feature extraction module and the prediction module. Optionally, in this embodiment, an SPP module is used as the feature fusion module, and a structure of FPN + PAN is used to perform parameter aggregation on different detection layers from different trunk layers of the deep convolutional neural network, so as to further improve the feature extraction capability. Finally, a plurality of feature maps with different scales are output.
And (4) feature fusion, which is to perform multi-scale feature fusion on the feature extraction performed in the feature extraction step S24, preferably, the feature fusion may use a path aggregation network to generate three-scale feature maps, which are respectively used for detecting objects of different sizes.
S26: obtaining a plurality of prediction results of different scales for a plurality of feature maps of different scales through a prediction module; the predicted result at least comprises: whether foreign matter exists in the sample data, the type of the foreign matter and the coordinate position of the foreign matter.
If the feature maps of three dimensions are generated in step S25, the prediction results of three dimensions are obtained, and the prediction results include whether a foreign object exists in the sample data, the type of the foreign object, and the coordinate position of the foreign object.
Whether foreign objects exist in the sample data is represented by confidence, and the coordinate positions of the foreign objects can be represented by the coordinates of the bounding box.
And in a final prediction module, screening a target frame by adopting a non-maximum suppression algorithm, and finally outputting the detected target.
In this embodiment, a CIOU loss function is adopted, and a calculation formula thereof is as follows:
wherein b is a candidate target frame, bgtMarking a target frame; where ρ is the distance between the center points of the prediction frame and the labeling frame, and c is the distance just enough to contain the measurement frameAnd the length of the diagonal of the smallest rectangle of the label box.
Preferably, in the tobacco shred foreign matter identification model in the embodiment, the model is lightened by methods such as pruning, calculation accuracy reduction, knowledge distillation and the like, so that certain accuracy is ensured and calculation complexity is reduced.
After the tobacco foreign matter identification model is trained, the tobacco foreign matter identification model is operated on line, the tobacco foreign matter identification model is deployed to a graphic workstation by using a tf-lite frame, a corresponding application program is programmed, and the tobacco foreign matter detection function is executed.
A tobacco foreign matter recognition apparatus, referring to fig. 3, comprising:
a processor 31 and a memory 32;
the processor 31 and the memory 32 are connected by a communication bus:
the processor 31 is used for calling and executing the program stored in the memory 32;
the memory 32 is used for storing a program, and the program is at least used for executing the cut tobacco foreign matter identification method in any one of the above embodiments.
The embodiment of the present invention may further include a storage medium, where the storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the tobacco shred foreign matter identification method are implemented.
It is understood that the same or similar parts in the above embodiments may be mutually referred to, and the same or similar parts in other embodiments may be referred to for the content which is not described in detail in some embodiments.
It should be noted that, in the description of the present application, the terms "first", "second", etc. are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Further, in the description of the present application, the meaning of "a plurality" means at least two unless otherwise specified.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and the scope of the preferred embodiments of the present application includes other implementations in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (10)
1. A tobacco shred foreign matter identification method is characterized by comprising the following steps:
acquiring a tobacco shred image acquired by an image acquisition device;
preprocessing the tobacco shred image;
obtaining a recognition result of the tobacco shred image based on a pre-trained tobacco shred foreign matter recognition model according to the preprocessed tobacco shred image; the recognition result at least comprises: whether foreign matters exist in the tobacco shred images or not, the types of the foreign matters and the coordinate positions of the foreign matters.
2. The method of claim 1, further comprising:
and acquiring a tobacco shred image doped with foreign matters as sample data, and training the tobacco shred foreign matter identification model.
3. The method according to claim 2, wherein the training of the tobacco shred foreign object recognition model specifically comprises:
marking the sample data based on a marking tool;
establishing the tobacco shred foreign matter identification model by adopting a deep neural network target detection framework;
inputting the marked sample data into the tobacco shred foreign matter identification model;
extracting the characteristics of the marked sample data based on the tobacco shred foreign matter identification model;
performing feature fusion on the extracted features based on the tobacco shred foreign matter identification model, and outputting a plurality of feature maps with different scales;
obtaining a plurality of prediction results of different scales for the feature maps of different scales through a prediction module; the prediction result at least comprises: and whether the sample data contains foreign matters, the types of the foreign matters and the coordinate positions of the foreign matters or not.
4. The method according to claim 3, wherein said labeling the sample data based on a labeling tool specifically comprises:
identifying foreign objects in the sample data by using a minimum outsourcing rectangular frame;
establishing a coordinate system by taking the sample data as a reference, and acquiring the coordinate of the central point of the foreign matter in the coordinate system;
acquiring the length and the width of the minimum external rectangular frame of the foreign matters;
identifying a type of the foreign matter.
5. The method according to claim 3, wherein before inputting the labeled sample data into the tobacco shred foreign matter identification model, the method further comprises:
and performing image enhancement on the labeled sample data based on image occlusion and multi-image combination modes.
6. The method according to claim 3, wherein before inputting the labeled sample data into the tobacco shred foreign matter identification model, the method further comprises:
and carrying out grid division on the labeled sample data.
7. The method according to claim 3, wherein the characteristic extraction of the labeled sample data based on the tobacco shred foreign matter identification model specifically comprises:
and dividing the feature mapping of the deep convolutional neural network basic layer into two parts, merging the divided two parts of feature mapping through a cross-stage hierarchical structure, and repeatedly executing for multiple times to obtain a feature map.
8. The method according to claim 3, wherein the extracting features are subjected to feature fusion based on the tobacco shred foreign matter identification model, and a plurality of feature maps with different scales are output, specifically comprising:
and carrying out parameter aggregation on different detection layers from different trunk layers of the deep convolutional neural network, improving the capability of feature extraction, and outputting a plurality of feature maps with different scales.
9. The method according to claim 1, wherein the preprocessing the cut tobacco image specifically comprises:
segmenting the tobacco shred image with high resolution;
and carrying out grid division on the cut tobacco image after segmentation.
10. A tobacco shred foreign matter recognition device is characterized by comprising:
a processor and a memory;
the processor and the memory are connected through a communication bus:
the processor is used for calling and executing the program stored in the memory;
the memory is used for storing a program at least for executing the cut tobacco foreign matter identification method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110160109.1A CN112819796A (en) | 2021-02-05 | 2021-02-05 | Tobacco shred foreign matter identification method and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110160109.1A CN112819796A (en) | 2021-02-05 | 2021-02-05 | Tobacco shred foreign matter identification method and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112819796A true CN112819796A (en) | 2021-05-18 |
Family
ID=75861612
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110160109.1A Pending CN112819796A (en) | 2021-02-05 | 2021-02-05 | Tobacco shred foreign matter identification method and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112819796A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469087A (en) * | 2021-07-09 | 2021-10-01 | 上海智臻智能网络科技股份有限公司 | Method, device, equipment and medium for detecting picture frame in building drawing |
| CN113887597A (en) * | 2021-09-24 | 2022-01-04 | 青岛浩海网络科技股份有限公司 | A smoke recognition method based on deep learning |
| CN113916899A (en) * | 2021-10-11 | 2022-01-11 | 四川科伦药业股份有限公司 | Method, system and device for detecting large soft infusion bag product based on visual identification |
| CN114005081A (en) * | 2021-09-24 | 2022-02-01 | 常州市新科汽车电子有限公司 | Intelligent detection device and method for foreign matters in tobacco shreds |
| CN114202516A (en) * | 2021-11-29 | 2022-03-18 | 上海联影医疗科技股份有限公司 | Foreign matter detection method and device, electronic equipment and storage medium |
| CN114271763A (en) * | 2021-12-20 | 2022-04-05 | 合肥中纳医学仪器有限公司 | Mask RCNN-based gastric cancer early identification method, system and device |
| CN114972800A (en) * | 2022-05-13 | 2022-08-30 | 河南中烟工业有限责任公司 | Metering trough foreign matter recognition early warning method |
| CN115713527A (en) * | 2022-11-27 | 2023-02-24 | 河南中烟工业有限责任公司 | Method and system for detecting foreign matters in cut tobacco |
| CN115984636A (en) * | 2023-03-21 | 2023-04-18 | 杭州书微信息科技有限公司 | Foreign matter impurity removal system and method |
| CN116309351A (en) * | 2023-02-15 | 2023-06-23 | 浙江丽威汽车控制系统有限公司 | Automobile engineering material supply processing system |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330882A (en) * | 2017-06-30 | 2017-11-07 | 航天新长征大道科技有限公司 | Foreign matter online test method after a kind of cut based on machine vision |
| CN109214336A (en) * | 2018-09-05 | 2019-01-15 | 高新兴科技集团股份有限公司 | A kind of vehicle window marker detection method and device |
| CN109297969A (en) * | 2018-12-05 | 2019-02-01 | 上海庚葵贸易有限公司 | A kind of tobacco leaf assembly line method for detecting impurities and system |
| CN109409365A (en) * | 2018-10-25 | 2019-03-01 | 江苏德劭信息科技有限公司 | It is a kind of that method is identified and positioned to fruit-picking based on depth targets detection |
| CN109446369A (en) * | 2018-09-28 | 2019-03-08 | 武汉中海庭数据技术有限公司 | The exchange method and system of the semi-automatic mark of image |
| CN109523563A (en) * | 2018-11-01 | 2019-03-26 | 华东师范大学 | A method of improving micro-image annotating efficiency |
| CN109766884A (en) * | 2018-12-26 | 2019-05-17 | 哈尔滨工程大学 | A Foreign Object Detection Method on Airport Runway Based on Faster-RCNN |
| CN110321891A (en) * | 2019-03-21 | 2019-10-11 | 长沙理工大学 | A kind of big infusion medical fluid foreign matter object detection method of combined depth neural network and clustering algorithm |
| CN110826416A (en) * | 2019-10-11 | 2020-02-21 | 佛山科学技术学院 | A method and device for detecting surface defects of bathroom ceramics based on deep learning |
| CN111754502A (en) * | 2020-06-30 | 2020-10-09 | 浙江工业大学 | A method of detecting surface defects of magnetic core based on Faster-RCNN algorithm based on multi-scale feature fusion |
| CN111858980A (en) * | 2020-07-28 | 2020-10-30 | 东北大学 | A method for identifying and cataloging cores in core disks |
| CN111881774A (en) * | 2020-07-07 | 2020-11-03 | 上海艾豚科技有限公司 | Method and system for identifying foreign matters in textile raw materials |
| CN111882579A (en) * | 2020-07-03 | 2020-11-03 | 湖南爱米家智能科技有限公司 | Large infusion foreign matter detection method, system, medium and equipment based on deep learning and target tracking |
| CN112189877A (en) * | 2020-10-13 | 2021-01-08 | 燕山大学 | On-line detection method for tobacco shred impurities in tobacco production line |
-
2021
- 2021-02-05 CN CN202110160109.1A patent/CN112819796A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330882A (en) * | 2017-06-30 | 2017-11-07 | 航天新长征大道科技有限公司 | Foreign matter online test method after a kind of cut based on machine vision |
| CN109214336A (en) * | 2018-09-05 | 2019-01-15 | 高新兴科技集团股份有限公司 | A kind of vehicle window marker detection method and device |
| CN109446369A (en) * | 2018-09-28 | 2019-03-08 | 武汉中海庭数据技术有限公司 | The exchange method and system of the semi-automatic mark of image |
| CN109409365A (en) * | 2018-10-25 | 2019-03-01 | 江苏德劭信息科技有限公司 | It is a kind of that method is identified and positioned to fruit-picking based on depth targets detection |
| CN109523563A (en) * | 2018-11-01 | 2019-03-26 | 华东师范大学 | A method of improving micro-image annotating efficiency |
| CN109297969A (en) * | 2018-12-05 | 2019-02-01 | 上海庚葵贸易有限公司 | A kind of tobacco leaf assembly line method for detecting impurities and system |
| CN109766884A (en) * | 2018-12-26 | 2019-05-17 | 哈尔滨工程大学 | A Foreign Object Detection Method on Airport Runway Based on Faster-RCNN |
| CN110321891A (en) * | 2019-03-21 | 2019-10-11 | 长沙理工大学 | A kind of big infusion medical fluid foreign matter object detection method of combined depth neural network and clustering algorithm |
| CN110826416A (en) * | 2019-10-11 | 2020-02-21 | 佛山科学技术学院 | A method and device for detecting surface defects of bathroom ceramics based on deep learning |
| CN111754502A (en) * | 2020-06-30 | 2020-10-09 | 浙江工业大学 | A method of detecting surface defects of magnetic core based on Faster-RCNN algorithm based on multi-scale feature fusion |
| CN111882579A (en) * | 2020-07-03 | 2020-11-03 | 湖南爱米家智能科技有限公司 | Large infusion foreign matter detection method, system, medium and equipment based on deep learning and target tracking |
| CN111881774A (en) * | 2020-07-07 | 2020-11-03 | 上海艾豚科技有限公司 | Method and system for identifying foreign matters in textile raw materials |
| CN111858980A (en) * | 2020-07-28 | 2020-10-30 | 东北大学 | A method for identifying and cataloging cores in core disks |
| CN112189877A (en) * | 2020-10-13 | 2021-01-08 | 燕山大学 | On-line detection method for tobacco shred impurities in tobacco production line |
Non-Patent Citations (3)
| Title |
|---|
| ALEXEY BOCHKOVSKIY 等: "YOLOv4: Optimal Speed and Accuracy of Object Detection", 《ARXIV:2004.10934V1 [CS.CV]》 * |
| 付航等: "基于智能相机的残烟丝异物智能剔除系统的研究", 《科技创新导报》 * |
| 祝荣壮: "一种烟丝精选装置系统设计", 《机械设计与制造工程》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469087A (en) * | 2021-07-09 | 2021-10-01 | 上海智臻智能网络科技股份有限公司 | Method, device, equipment and medium for detecting picture frame in building drawing |
| CN113887597A (en) * | 2021-09-24 | 2022-01-04 | 青岛浩海网络科技股份有限公司 | A smoke recognition method based on deep learning |
| CN114005081A (en) * | 2021-09-24 | 2022-02-01 | 常州市新科汽车电子有限公司 | Intelligent detection device and method for foreign matters in tobacco shreds |
| CN113916899B (en) * | 2021-10-11 | 2024-04-19 | 四川科伦药业股份有限公司 | Method, system and device for detecting large transfusion soft bag product based on visual identification |
| CN113916899A (en) * | 2021-10-11 | 2022-01-11 | 四川科伦药业股份有限公司 | Method, system and device for detecting large soft infusion bag product based on visual identification |
| CN114202516A (en) * | 2021-11-29 | 2022-03-18 | 上海联影医疗科技股份有限公司 | Foreign matter detection method and device, electronic equipment and storage medium |
| CN114271763A (en) * | 2021-12-20 | 2022-04-05 | 合肥中纳医学仪器有限公司 | Mask RCNN-based gastric cancer early identification method, system and device |
| CN114271763B (en) * | 2021-12-20 | 2024-05-28 | 合肥中纳医学仪器有限公司 | Mask RCNN-based gastric cancer early-stage identification method, system and device |
| CN114972800A (en) * | 2022-05-13 | 2022-08-30 | 河南中烟工业有限责任公司 | Metering trough foreign matter recognition early warning method |
| CN115713527A (en) * | 2022-11-27 | 2023-02-24 | 河南中烟工业有限责任公司 | Method and system for detecting foreign matters in cut tobacco |
| CN116309351A (en) * | 2023-02-15 | 2023-06-23 | 浙江丽威汽车控制系统有限公司 | Automobile engineering material supply processing system |
| CN116309351B (en) * | 2023-02-15 | 2023-11-21 | 浙江丽威汽车控制系统有限公司 | Automobile engineering material supply processing system |
| CN115984636A (en) * | 2023-03-21 | 2023-04-18 | 杭州书微信息科技有限公司 | Foreign matter impurity removal system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112819796A (en) | Tobacco shred foreign matter identification method and equipment | |
| US11276162B2 (en) | Surface defect identification method and apparatus | |
| CN112348787B (en) | Training method of object defect detection model, object defect detection method and device | |
| CN111027547A (en) | Automatic detection method for multi-scale polymorphic target in two-dimensional image | |
| CN111681273B (en) | Image segmentation method and device, electronic equipment and readable storage medium | |
| CN114049356B (en) | Method, device and system for detecting structure apparent crack | |
| KR101932009B1 (en) | Image processing apparatus and method for multiple object detection | |
| CN110490212A (en) | Molybdenum target image processing arrangement, method and apparatus | |
| CN112990085B (en) | Method and device for detecting change of culture pond and computer readable storage medium | |
| CN112819753B (en) | Building change detection method and device, intelligent terminal and storage medium | |
| CN114140665A (en) | A Dense Small Object Detection Method Based on Improved YOLOv5 | |
| CN114332473B (en) | Object detection method, device, computer apparatus, storage medium, and program product | |
| CN115019181B (en) | Remote sensing image rotating target detection method, electronic equipment and storage medium | |
| CN115272887A (en) | Coastal garbage identification method, device and equipment based on UAV detection | |
| CN112419202A (en) | Automatic recognition system of wild animal images based on big data and deep learning | |
| JP2024529423A (en) | Security inspection CT object recognition method and device | |
| CN109657540B (en) | Dead tree location method and system | |
| CN115439654A (en) | Method and system for finely dividing weakly supervised farmland plots under dynamic constraint | |
| CN116342525A (en) | SOP chip pin defect detection method and system based on Lenet-5 model | |
| CN112215217B (en) | Digital image recognition method and device for simulating doctor to read film | |
| CN115239644A (en) | Concrete defect identification method and device, computer equipment and storage medium | |
| CN116433596A (en) | Slope vegetation coverage measuring method and device and related components | |
| CN114792300B (en) | X-ray broken needle detection method based on multi-scale attention | |
| CN117456287B (en) | Method for observing population number of wild animals by using remote sensing image | |
| CN116895022B (en) | Building boundary extraction method based on point cloud data processing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210518 |
|
| RJ01 | Rejection of invention patent application after publication |