CN116778479A - Improved YOLOv 5-based worm egg detection method, device, equipment and medium - Google Patents
Improved YOLOv 5-based worm egg detection method, device, equipment and medium Download PDFInfo
- Publication number
- CN116778479A CN116778479A CN202310499396.8A CN202310499396A CN116778479A CN 116778479 A CN116778479 A CN 116778479A CN 202310499396 A CN202310499396 A CN 202310499396A CN 116778479 A CN116778479 A CN 116778479A
- Authority
- CN
- China
- Prior art keywords
- worm egg
- worm
- improved
- image
- detection
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 85
- 235000013601 eggs Nutrition 0.000 claims abstract description 166
- 238000000034 method Methods 0.000 claims abstract description 17
- 238000003384 imaging method Methods 0.000 claims abstract description 11
- 238000012549 training Methods 0.000 claims description 30
- 238000002372 labelling Methods 0.000 claims description 9
- 230000007246 mechanism Effects 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 5
- 230000003014 reinforcing effect Effects 0.000 claims description 3
- 210000001035 gastrointestinal tract Anatomy 0.000 description 17
- 238000000605 extraction Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 3
- 239000007788 liquid Substances 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 208000009849 Female Genital Neoplasms Diseases 0.000 description 1
- 208000030852 Parasitic disease Diseases 0.000 description 1
- FAPWRFPIFSIZLT-UHFFFAOYSA-M Sodium chloride Chemical compound [Na+].[Cl-] FAPWRFPIFSIZLT-UHFFFAOYSA-M 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000005314 correlation function Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000007865 diluting Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 244000005700 microbiome Species 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 210000003097 mucus Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 244000045947 parasite Species 0.000 description 1
- 230000002572 peristaltic effect Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000003761 preservation solution Substances 0.000 description 1
- 239000011780 sodium chloride Substances 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/698—Matching; Classification
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/766—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using regression, e.g. by projecting features on hyperplanes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/12—Acquisition of 3D measurements of objects
- G06V2201/122—Computational image acquisition in electron microscopy
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a worm egg detection method, a device, electronic equipment and a storage medium based on improved YOLOv5, wherein the method comprises the following steps: obtaining a worm egg image dataset based on a full-automatic microscope imaging system; and inputting the worm egg image data set into an improved YOLOv5 model to detect worm eggs, and obtaining a worm egg class detection result. The application solves the technical problems of low detection efficiency and low detection precision of vermicular eggs caused by low automatic detection degree of vermicular eggs and dependence on professional reader in the prior art.
Description
Technical Field
The application relates to the technical field of medical detection, in particular to a worm egg detection method, device, equipment and medium based on improved YOLOv 5.
Background
The following methods are commonly used in the art for gut worm detection: (1) saline smear method, mild parasitic infection is not easily detected by direct smear method, especially when the number of parasites is small. The method shows the peristaltic movement of the microorganisms through a uniform suspension. (2) The liquid-based worm egg enrichment method is a conventional screening technology all the time, and the liquid-based cytology principle adopted is completely the same as that adopted for gynecological and non-gynecological tumor diagnosis. And diluting the sample in a preservation solution, removing other influencing factors such as mucus and the like, and placing the sample on a glass slide for microscopic observation and analysis. (3) The morphological detection module of the stool analyzer uses a microscope to observe microscopic morphology of specimen application liquid, and a full-automatic microscope system (automatically acquiring pictures for operators to examine) and a quasi-automatic microscope system are common, so that manual real-time observation and selective photographing are performed. The professional requirements for the reader are high, and the positive detection rate is relatively low.
Disclosure of Invention
The application aims to overcome the technical defects, and provides a worm egg detection method, device, electronic equipment and storage medium based on improved YOLOv5, which solve the technical problems of low detection efficiency and detection accuracy caused by low automation level of worm egg detection and dependence on professional film readers in the prior art.
In order to achieve the technical purpose, the application adopts the following technical scheme:
in a first aspect, the present application provides an improved YOLOv 5-based worm egg detection method comprising:
obtaining a worm egg image dataset based on a full-automatic microscope imaging system;
inputting the worm egg image data set into an improved YOLOv5 model for worm egg detection to obtain a worm egg class detection result;
wherein the improved YOLOv5 model comprises an effective channel attention mechanism module for extracting a vermicular worm image feature map.
In some embodiments, the inputting the worm egg image dataset into the modified YOLOv5 model for worm egg detection further comprises:
labeling the collected worm egg image to obtain a labeled worm egg image data set;
dividing the marked vermicular worm egg image data set into a worm egg image test set and a worm egg image training set;
inputting the worm egg image training set into an improved YOLOv5 model, and determining a worm egg image prediction set corresponding to the worm egg image training set, wherein the effective channel attention mechanism module is used for extracting feature images of different scales of worm egg images;
and determining a value of a loss function according to the error between the worm egg image training set and the worm egg image prediction set, and adjusting parameters of the improved YOLOv5 model according to the value of the loss function until convergence conditions are met, so as to determine the improved YOLOv5 model with complete training.
In some embodiments, before the inputting the worm egg image training set into the improved YOLOv5 model, further comprising:
reinforcing the vermicular egg worms by a preset mosaics method to obtain reinforced vermicular egg worm images;
and slicing the enhanced vermicular worm image by adopting a preset size, and determining the vermicular worm image training sample set.
In some embodiments, the labeling the collected worm egg image to obtain a labeled worm egg image dataset comprises:
labeling the worm egg image based on preset LabelImg open source software to obtain a labeled worm egg image; the noted worm egg image includes worm egg rectangular bounding box coordinates.
In some embodiments, the inputting the worm egg image dataset into the improved YOLOv5 model for worm egg detection, after obtaining the detection result of the vermicular egg class, further comprises:
and based on the detection result, evaluating the improved YOLOv5 model with complete training by adopting a preset average precision and average precision mean value, and determining the detection precision of the improved YOLOv5 model.
In some embodiments, the preset average precision may be expressed by the following formula:
wherein TP (True positive) is the number of positive samples detected correctly, FP (False positive) is the number of positive samples detected incorrectly.
In some embodiments, the preset average precision mean may be expressed by the following formula:
wherein TP (True positive) is the number of positive samples detected correctly, i.e. the predicted frame and the labeled frame are the same in category and IoU >0.5; FP (False positive) is the number of positive samples in which an error is detected; FN is the number of negative samples in which errors are detected; r is the whole real number set; AP is the area under the recall and precision curves.
In a second aspect, the present application also provides an improved YOLOv 5-based worm egg detection device, comprising:
the obtaining module is used for obtaining a worm egg image data set based on the full-automatic microscope imaging system;
the detection module is used for inputting the worm egg image data set into an improved YOLOv5 model to detect worm eggs and obtain worm egg class detection results.
In a third aspect, the present application also provides an electronic device, including: a processor and a memory;
the memory has stored thereon a computer readable program executable by the processor;
the processor, when executing the computer readable program, implements the steps in the improved YOLOv 5-based worm egg detection method as described above.
In a fourth aspect, the present application also provides a computer readable storage medium storing one or more programs executable by one or more processors to implement the steps in the improved YOLOv 5-based worm egg detection method as described above.
Compared with the prior art, the worm egg detection method, the device, the electronic equipment and the storage medium based on the improved YOLOv5 provided by the application are characterized in that firstly, a worm egg image data set is obtained based on a full-automatic microscope imaging system; and finally, inputting the worm egg image data set into an improved YOLOv5 model to detect worm eggs, and obtaining a worm egg type detection result. According to the application, the target vermicular worm eggs are optimized and improved according to the characteristics of dense distribution and smaller size, the automatic target detection precision is improved, the standardized sampling and the auxiliary diagnosis of the vermicular worm eggs in the digestive tract are realized, the professional requirements of the reader are reduced, the positive detection rate is improved, and the accuracy of screening the vermicular worm diseases in the digestive tract is improved.
Drawings
FIG. 1 is a flow chart of one embodiment of the improved Yolov 5-based worm egg detection method provided by the present application;
FIG. 2 is a schematic diagram of an embodiment of an improved Yolov model training procedure in the improved Yolov 5-based worm egg detection method of the present application;
FIG. 3 is a schematic diagram of an embodiment of worm egg image enhancement in the improved Yolov 5-based worm egg detection method provided by the present application;
FIG. 4 is a schematic diagram of an embodiment of the improved Yolov 5-based worm egg detection apparatus provided by the present application;
FIG. 5 is a schematic diagram of an operating environment of an embodiment of an electronic device according to the present application.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
Aiming at the problems of small and densely distributed digestive tract worm egg targets, low automation degree, low efficiency and the like in microscopic digestive tract worm egg images acquired by a microscope, the application provides a digestive tract worm egg target detection method based on improved YOLOv5 based on a deep learning method, so as to realize rapid, accurate and automatic identification and positioning of digestive tract worm eggs.
The embodiment of the application provides a worm egg detection method based on improved YOLOv5, referring to fig. 1, comprising the following steps:
s101, obtaining a worm egg image data set based on a full-automatic microscope imaging system;
s102, inputting the worm egg image dataset into an improved YOLOv5 model for worm egg detection to obtain a worm egg class detection result;
wherein the improved YOLOv5 model comprises an effective channel attention mechanism module for extracting a vermicular worm image feature map.
In the embodiment, a worm egg image dataset is obtained firstly based on a full-automatic microscope imaging system; and finally, inputting the worm egg image data set into an improved YOLOv5 model to detect worm eggs, and obtaining a worm egg type detection result. According to the application, the target vermicular worm eggs are optimized and improved according to the characteristics of dense distribution and smaller size, the automatic target detection precision is improved, the standardized sampling and the auxiliary diagnosis of the vermicular worm eggs in the digestive tract are realized, the professional requirements of the reader are reduced, the positive detection rate is improved, and the accuracy of screening the vermicular worm diseases in the digestive tract is improved.
In step S101, a full-automatic microscope imaging system is used to collect raw sheets with 23184 x 16557 resolution of vermicular worm eggs in the digestive tract, and then the raw sheets are divided into 13 images with 1920 x 1200 resolution in a mode of overlapping adjacent image boundaries; thus, the tiny vermicular worm eggs in the digestive tract can be displayed by a 20-time or 40-time microscope in a graphical manner. The algorithm of YOLOv5 was applied to the ova of the micro-molecules.
In step S102, YOLOv5 is selected as a basic target detection model, and is optimized and improved according to the characteristics of dense distribution and smaller size of the worm eggs in the digestive tract, so as to improve the automatic target detection precision. And (3) through a training set of the vermicular worm eggs in the alimentary canal, a worm egg target is identified by reasoning operation, and a target result is returned in json format.
In some embodiments, referring to fig. 2, before the step of inputting the worm egg image dataset into the modified YOLOv5 model for worm egg detection, the method further comprises:
s201, marking the collected worm egg image to obtain a marked worm egg image data set;
s202, dividing the marked vermicular worm egg image data set into a worm egg image test set and a worm egg image training set;
s203, inputting the worm egg image training set into an improved YOLOv5 model, and determining a worm egg image prediction set corresponding to the worm egg image training set, wherein the improved YOLOv5 model comprises an effective channel attention mechanism module, and the effective channel attention mechanism module is used for extracting feature images of different scales of a worm image;
s204, determining a value of a loss function according to the error between the worm egg image training set and the worm egg image prediction set, and adjusting parameters of the improved YOLOv5 model according to the value of the loss function until convergence conditions are met, so as to determine the improved YOLOv5 model with complete training.
In this example, since the eggs of the digestive tract worms are densely distributed and the targets are small, they are micro-cells, and the data are obtained by means of a 20-fold or 40-fold microscope imaging system to construct the required data set; introducing an effective channel attention mechanism (Efficient Channel Attention) module in a CSPDarknet53 feature extraction network of the Yolov5, and improving the Yolov5 network; inputting the obtained images of the digestive tract worm egg data set into a Yolov5 feature extraction network for feature extraction to obtain feature images with different scales; classifying and regressing the obtained feature images, performing feature reconstruction operation on the regressing result to obtain finer feature images, performing classification and regressing operation again on the basis, and calculating loss; the training accuracy of the model can be improved.
Further, the effective channel attention mechanism (Efficient Channel Attention) module generates channel attention through a fast one-dimensional convolution with a size k, wherein the size of a kernel is determined by channel dimension correlation function adaptation, a feature image χ is input under the condition that the dimension is kept unchanged, after all channels are subjected to global average pooling, the ECA module learns features through a one-dimensional convolution which can be shared by weights, and when learning features, each channel is involved in capturing cross-channel interaction with k neighbors thereof, k represents the kernel size of the fast one-dimensional convolution, and the determination of the adaptive k value is obtained through the proportional relation between a coverage area of the cross-channel information interaction and the channel dimension C, and the calculation is as shown in a formula (1):
wherein: γ=2, b=1, | ood denotes the nearest odd number and C is the channel dimension.
In some embodiments, referring to fig. 3, before inputting the worm egg image training set into the modified YOLOv5 model, the method further comprises:
s301, reinforcing the vermicular egg worm by adopting a preset mosaics method to obtain a reinforced vermicular egg worm image;
s302, slicing the enhanced vermicular worm image by adopting a preset size, and determining the vermicular worm image training sample set.
In this embodiment, before performing feature extraction on the Yolov5 feature extraction network, performing a mosaic enhancement operation on the data, uniformly scaling the data to a standard size to perform a Focus slicing operation, and inputting the data to the Yolov5 feature extraction network to perform feature extraction, so that the learning capability of a sample can be enhanced, and the detection accuracy of a model can be improved.
In some embodiments, the labeling the collected worm egg image to obtain a labeled worm egg image dataset comprises:
labeling the worm egg image based on preset LabelImg open source software to obtain a labeled worm egg image; the noted worm egg image includes worm egg rectangular bounding box coordinates.
In this embodiment, in order to ensure the integrity of each detection target in the image, labelImg open source software is used to label the digestive tract worm eggs in turn, the label content is 8 kinds of rectangular bounding box coordinates of the digestive tract worm eggs, and the rectangular bounding box coordinates are stored as TXT text files, which are used for training and testing the YOLOv5 model, the YOLOv5 structure includes an Input end Input, a reference network back, a feature fusion part negk and a detection Head part, the Input end of the YOLOv5 includes a preprocessing stage of a training data set image, the Mosaic data of the YOLOv5 is used to enhance the training speed and network precision of the model, and an adaptive anchor frame calculation program is newly added, a Focus structure is added in the Backbone network part, and used to extract general features, and two kinds of structures are constructed according to the Focus structure, the image is sliced, a feature fusion part in the middle of the Backbone network and the Head detection is written, the YOLOv5 adds a combined structure of fpn+pan, and simultaneously cuts the image correspondingly and cuts the image and supplements the required target on the digestive tract with the complete target, and the label is not required to be labeled on the target.
In some embodiments, the inputting the worm egg image dataset into the improved YOLOv5 model for worm egg detection, after obtaining the detection result of the vermicular egg class, further comprises:
and based on the detection result, evaluating the improved YOLOv5 model with complete training by adopting a preset average precision and average precision mean value, and determining the detection precision of the improved YOLOv5 model.
In this embodiment, the test set is tested by means of the divided data set to achieve target detection of the vermicular worm eggs of the alimentary canal, and the detection effect of the model is evaluated.
The detection effect and performance of the model are generally evaluated by using the average mAP of the average accuracy AP (Average Precision) and the average accuracy, and AP is the area under the Recall ratio Recall and the accuracy Precision curve, and the area intersection ratio IoU: and calculating the area intersection ratio of the rectangular area of the target predicted by the model and the rectangular area calibrated by the target in the verification set, and measuring the position prediction capability of the model.
Accuracy rate Precision, which is to represent the proportion of the correct target number to the total target number detected by a model, and to represent the accuracy of the model in target detection, is generally expressed by the following formula:
recall ratio (Recall) the Recall ratio represents the ratio of the number of targets detected by the model to the total number of targets, and represents the full searching capability of model identification.
Wherein: TP (True positive) is the number of positive samples detected correctly, i.e. the predicted frame and the labeled frame are the same in category and IoU >0.5; FP (False positive) is the number of positive samples in which an error is detected; FN is the number of negative samples in which errors are detected; r is the whole real number set; AP is the area under the recall and precision curves.
Based on the above-mentioned method for detecting worm eggs based on improved YOLOv5, the embodiment of the present application further provides a device 400 for detecting worm eggs based on improved YOLOv5, please refer to fig. 4, and the device 400 for detecting worm eggs based on improved YOLOv5 includes an acquisition module 410 and a detection module 420.
An acquisition module 410 for acquiring a worm egg image dataset based on a fully automated microscope imaging system;
the detection module 420 is configured to input the worm egg image dataset into an improved YOLOv5 model for worm egg detection, and obtain a worm egg class detection result.
As shown in fig. 5, based on the above-mentioned worm egg detection method based on improved YOLOv5, the present application further provides an electronic device, which may be a mobile terminal, a desktop computer, a notebook computer, a palm computer, a server, and other computing devices. The electronic device includes a processor 510, a memory 520, and a display 530. Fig. 5 shows only some of the components of the electronic device, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead.
The memory 520 may be an internal storage unit of the electronic device, such as a hard disk or memory of the electronic device, in some embodiments. The memory 520 may also be an external storage device of the electronic device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like. Further, the memory 520 may also include both internal storage units and external storage devices of the electronic device. The memory 520 is used for storing application software installed on the electronic device and various data, such as program codes for installing the electronic device. The memory 520 may also be used to temporarily store data that has been output or is to be output. In one embodiment, the memory 520 has stored thereon an improved YOLOv 5-based worm egg detection program 540, the improved YOLOv 5-based worm egg detection program 540 being executable by the processor 510 to implement the improved YOLOv 5-based worm egg detection method of embodiments of the present application.
The processor 510 may in some embodiments be a central processing unit (Central Processing Unit, CPU), microprocessor or other data processing chip for running program code or processing data stored in the memory 20, e.g. performing the improved YOLOv 5-based worm egg detection method, etc.
The display 530 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like in some embodiments. Display 530 is used to display information at the improved YOLOv 5-based worm egg detection device and to display a visual user interface. The components 510-530 of the electronic device communicate with each other over a system bus.
Of course, those skilled in the art will appreciate that implementing all or part of the above-described methods may be implemented by a computer program for instructing relevant hardware (e.g., a processor, a controller, etc.), where the program may be stored in a computer-readable storage medium, and where the program may include the steps of the above-described method embodiments when executed. The storage medium may be a memory, a magnetic disk, an optical disk, or the like.
The above-described embodiments of the present application do not limit the scope of the present application. Any other corresponding changes and modifications made in accordance with the technical idea of the present application shall be included in the scope of the claims of the present application.
Claims (10)
1. A method for detecting worm eggs based on improved YOLOv5, comprising:
obtaining a worm egg image dataset based on a full-automatic microscope imaging system;
inputting the worm egg image data set into an improved YOLOv5 model for worm egg detection to obtain a worm egg class detection result;
wherein the improved YOLOv5 model comprises an effective channel attention mechanism module for extracting a vermicular worm image feature map.
2. The improved YOLOv 5-based worm egg detection method of claim 1, wherein the inputting of worm egg image datasets into an improved YOLOv5 model for worm egg detection is preceded by:
labeling the collected worm egg image to obtain a labeled worm egg image data set;
dividing the marked vermicular worm egg image data set into a worm egg image test set and a worm egg image training set;
inputting the worm egg image training set into an improved YOLOv5 model, and determining a worm egg image prediction set corresponding to the worm egg image training set, wherein the effective channel attention mechanism module is used for extracting feature images of different scales of worm egg images;
and determining a value of a loss function according to the error between the worm egg image training set and the worm egg image prediction set, and adjusting parameters of the improved YOLOv5 model according to the value of the loss function until convergence conditions are met, so as to determine the improved YOLOv5 model with complete training.
3. The improved YOLOv 5-based worm egg detection method of claim 2, further comprising, prior to inputting the worm egg image training set into an improved YOLOv5 model:
reinforcing the vermicular egg worms by a preset mosaics method to obtain reinforced vermicular egg worm images;
and slicing the enhanced vermicular worm image by adopting a preset size, and determining the vermicular worm image training sample set.
4. The improved YOLOv 5-based worm egg detection method of claim 2, wherein labeling the acquired worm egg image to obtain a labeled worm egg image dataset comprises:
labeling the worm egg image based on preset LabelImg open source software to obtain a labeled worm egg image; the noted worm egg image includes worm egg rectangular bounding box coordinates.
5. The improved YOLOv 5-based worm egg detection method of claim 1, wherein the step of inputting the worm egg image dataset into an improved YOLOv5 model for worm egg detection, after obtaining a worm egg class detection result, further comprises:
and based on the detection result, evaluating the improved YOLOv5 model with complete training by adopting a preset average precision and average precision mean value, and determining the detection precision of the improved YOLOv5 model.
6. The improved YOLOv 5-based worm egg detection method of claim 1, wherein the predetermined average accuracy is expressed by the following formula:
wherein TP (True positive) is the number of positive samples detected correctly, FP (False positive) is the number of positive samples detected incorrectly.
7. The improved YOLOv 5-based worm egg detection method of claim 1, wherein the preset average precision mean value is expressed by the following formula:
the TP is the number of detected correct positive samples, namely the types of the prediction frame and the labeling frame are the same and IoU is more than 0.5; FP is the number of positive samples in which an error is detected; FN is the number of negative samples in which errors are detected; r is the whole real number set; AP is the area under the recall and precision curves.
8. An improved YOLOv 5-based worm egg detection device, comprising:
the obtaining module is used for obtaining a worm egg image data set based on the full-automatic microscope imaging system;
the detection module is used for inputting the worm egg image data set into an improved YOLOv5 model to detect worm eggs and obtain worm egg class detection results.
9. An electronic device, comprising: a processor and a memory;
the memory has stored thereon a computer readable program executable by the processor;
the processor, when executing the computer readable program, implements the steps in the improved YOLOv 5-based worm egg detection method as defined in claims 1-7.
10. A computer readable storage medium storing one or more programs executable by one or more processors to perform the steps in the improved YOLOv 5-based worm egg detection method of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310499396.8A CN116778479A (en) | 2023-05-05 | 2023-05-05 | Improved YOLOv 5-based worm egg detection method, device, equipment and medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310499396.8A CN116778479A (en) | 2023-05-05 | 2023-05-05 | Improved YOLOv 5-based worm egg detection method, device, equipment and medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116778479A true CN116778479A (en) | 2023-09-19 |
Family
ID=88012318
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310499396.8A Pending CN116778479A (en) | 2023-05-05 | 2023-05-05 | Improved YOLOv 5-based worm egg detection method, device, equipment and medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116778479A (en) |
-
2023
- 2023-05-05 CN CN202310499396.8A patent/CN116778479A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11176676B2 (en) | Systems and methods for processing electronic images for computational detection methods | |
CN109886928B (en) | Target cell marking method, device, storage medium and terminal equipment | |
CN110245657B (en) | Pathological image similarity detection method and detection device | |
JP2020502534A (en) | Computer scoring based on primary staining and immunohistochemical imaging | |
CN111709421B (en) | Bird identification method, bird identification device, computer equipment and storage medium | |
US20230186659A1 (en) | Machine learning models for cell localization and classification learned using repel coding | |
US12008750B2 (en) | Generating annotation data of tissue images | |
CN113393454A (en) | Method and device for segmenting pathological target examples in biopsy tissues | |
WO2023121846A1 (en) | Adversarial robustness of deep learning models in digital pathology | |
CN116205918A (en) | Multi-mode fusion semiconductor detection method, device and medium based on graph convolution | |
CN116778479A (en) | Improved YOLOv 5-based worm egg detection method, device, equipment and medium | |
CN116057585A (en) | Active learning system for digital pathology | |
CN112084889A (en) | Image behavior recognition method and device, computing equipment and storage medium | |
US20230024468A1 (en) | Systems and methods to process electronic images to determine histopathology quality | |
US20240221360A1 (en) | Machine-learning techniques for predicting phenotypes in duplex digital pathology images | |
CN117576109B (en) | Defect detection method, device, equipment and storage medium | |
US20230115448A1 (en) | Systems and methods for processing electronic images with preanalytic adjustment | |
WO2023043709A1 (en) | Machine-learning techniques for predicting phenotypes in duplex digital pathology images | |
WO2024118523A1 (en) | Consensus labeling in digital pathology images | |
CN117992626A (en) | Chromosome scanning synchronization method and system | |
CN117355870A (en) | System and method for processing electronic images to classify tissue type of specimen within a slide | |
CN116868229A (en) | Systems and methods for biomarker detection in digitized pathology samples | |
CN118135333A (en) | Medical picture intelligent sorting method and device, electronic equipment and readable storage medium | |
CN116503338A (en) | Thyroid cell pathology whole-slide image analysis method based on target detection | |
CN116129105A (en) | Target object position relation detection method, computing equipment and storage medium |
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 |