CN112115982A - Yoov 3-based automatic detection method for road surface leakage diseases - Google Patents
Yoov 3-based automatic detection method for road surface leakage diseases Download PDFInfo
- Publication number
- CN112115982A CN112115982A CN202010875105.7A CN202010875105A CN112115982A CN 112115982 A CN112115982 A CN 112115982A CN 202010875105 A CN202010875105 A CN 202010875105A CN 112115982 A CN112115982 A CN 112115982A
- Authority
- CN
- China
- Prior art keywords
- pavement
- leakage
- sample set
- detection
- leakage disease
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
- G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06N3/08—Learning methods
-
- 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
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a yolov 3-based automatic detection method for pavement leakage diseases, which relates to the technical field of image processing, and comprises the steps of collecting pavement leakage disease images to form a source sample set, and preprocessing the source sample set to obtain a training sample set; training a yolov3 target detection network by using the training sample set to obtain an initial pavement leakage disease detector; deploying the initial pavement leakage disease detector on a target road section to detect pavement leakage diseases, and recording a detection result; clustering and manually intervening the detection results, then labeling to obtain a target sample set, and carrying out online fine adjustment on the initial pavement leakage disease detector by using the target sample set, so that the accuracy of the initial pavement leakage disease detector on the target road section is improved, and the target pavement leakage disease detector is obtained; the iterative training method adopted by the invention has good generalization capability and can improve the identification precision of the unseen road scene and the unseen road leakage disease.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a pavement leakage disease automatic detection method based on yolov 3.
Background
Road appearance construction is being developed in China with great land to be fired, road cleaning quality supervision is promoted through intelligent detection of road leakage diseases, and the method has important significance for promoting intelligent maintenance and cleaning intelligence. At present, the treatment market of domestic garbage in China can reach 1100 hundred million per year, and the road cleaning market can reach about 550 hundred million per year. The intelligent identification of the pavement leakage diseases can assist the road management department to reasonably plan and design a road cleaning management mode, improve the quality of sanitation cleaning operation and effectively reduce the road cleaning management cost.
At present, the intelligent garbage recognition algorithm based on computer vision mainly has two types: the first is a convolutional neural network model for improving CaffeNet in the application of improved CaffeNet model in water surface garbage recognition, which is proposed in the document to improve the accuracy of water surface garbage recognition. The model improves the size of convolution kernels and the number of the convolution kernels, and increases a layer of sparse structure, so that the capability of extracting the characteristics of the network model is enhanced, and the complexity of the network is reduced. The method can reduce the influence of water surface ripples, object reflection, bridges and the like on water surface garbage recognition, and has a good water surface garbage recognition effect. But the robustness of the method is poor, and the method is difficult to be directly applied to more complex environments such as a road surface and the like; the garbage recognition and detection algorithm based on SSD, which is proposed in the 'garbage recognition classification research based on SSD algorithm', utilizes data enhancement to improve the robustness of a model, can achieve the purpose of quickly and accurately recognizing different types of garbage, and the method has good real-time performance but insufficient recognition capability on small targets.
It can be seen that, for the detection of small targets, the current common method at home and abroad is to use a multi-scale image pyramid or to upsample the small targets into large targets for detection. However, these two processing methods will seriously affect the time efficiency of the algorithm, thereby making it difficult to deploy and run the model on the edge computing device with low computational power.
Another problem is that the object to be identified must generally have a more uniform and invariant visual characteristic. The environment on the road surface is complex, and the road surface leakage disease is a fuzzy definition which is difficult to be solved. They tend to have visual characteristics that are closely related (e.g., visual characteristics of wood chips and leaves are close) and have large intra-class differences (e.g., plastic bags of different sizes and colors are very different). Therefore, the existing target detection algorithm and the deployment and use mode thereof can bring the problems of missing detection and false detection.
Therefore, the technical personnel in the field are dedicated to developing a pavement leakage disease automatic detection method based on yolov3, overcoming the problem of insufficient generalization capability of the prior art, and aiming at different pavement scenes and newly appeared pavement leakage diseases, the method can be self-adaptively adjusted and trained, and the accuracy of leakage disease identification is gradually improved.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a yolov 3-based automatic detection method for pavement leakage diseases, which can solve the problems that when the number of the pavement leakage diseases is large and most of the pavement leakage diseases are small targets, the recognition accuracy and the time efficiency cannot be simultaneously considered, and the problems of false detection and missed detection of an algorithm facing to a new road environment and the leakage diseases which do not appear in training data.
In order to achieve the purpose, the invention discloses a pavement leakage disease automatic detection method based on yolov3, which comprises the following steps:
s1, collecting road surface leakage disease images to form a source sample set, and preprocessing the source sample set to obtain a training sample set;
s2, training a yolov3 target detection network by utilizing a training sample set to obtain an initial pavement leakage disease detector;
s3, deploying an initial pavement leakage disease detector on the target road section to detect the pavement leakage diseases, and recording the detection result;
s4, clustering and manually intervening the detection results and then labeling to obtain a target sample set, and carrying out online fine adjustment on the initial pavement leakage disease detector by using the target sample set, so that the accuracy of the target road section is improved, and the target pavement leakage disease detector is obtained.
Preferably, S1 further includes:
counting the occurrence frequency of various broadcast leakage diseases in the source sample set, sequencing, and selecting the first N categories with the highest occurrence frequency as N broadcast leakage diseases to be detected;
and labeling each sample in the training sample set to obtain a sample label corresponding to the training sample set.
Preferably, the pre-treatment comprises:
horizontally turning the image to obtain an image to be further processed;
randomly carrying out translation transformation, rotation transformation and color transformation on the image to be further processed to obtain an image to be normalized;
and carrying out normalization processing on the image to be normalized.
Preferably, S2 further includes:
s201, training a yolov3 target detection network by adopting a random gradient descent algorithm, setting the maximum iteration times, calculating a network error by adopting a minimum batch mode (mini-batch) in each iteration, and updating parameters in the yolov3 target detection network by using the network error;
s202, when the preset maximum iteration times are reached or the error rate on the verification set is not reduced any more, the training is stopped, and the initial pavement leakage disease detector is obtained.
Preferably, the network error is the error of the classification label and the sample label output by the yolov3 target detection network.
Preferably, S3 includes:
s301, collecting the road surface leakage disease image again to serve as an original detection image;
s302, preprocessing an original detection image to obtain a detection image, and forming a detection sample set;
s303, extracting the characteristics of the detected image by using a backbone network in the initial leakage disease detector, and extracting all levels of characteristics of the network to form a characteristic pyramid;
s304, feature classification;
s305, drawing a detection result of the broadcast leakage disease in the detection image, outputting the result and finishing the broadcast leakage disease detection of the current image;
s306, storing the original detection image and the detection result.
Preferably, the feature classification is to classify the features by using a classification network in the initial broadcast leakage disease detector, set a threshold, and determine that the image contains the broadcast leakage disease to be detected if the output value is greater than the threshold, or determine that the image contains the background.
Preferably, the detection result is the coordinates and confidence of the target output by the initial pavement leakage defect detector.
Preferably, S4 includes:
s401, setting a confidence threshold, traversing the confidence, directly adding the samples with the confidence greater than the confidence threshold into a training sample set, and adding the samples with the confidence less than the confidence threshold into a post-processing sample set;
s402, re-labeling the targets in the post-processing sample set, and adding the targets into the training sample set to obtain a target sample set;
and S403, repeating S201 and S202 on the initial pavement leakage disease detector by using the target sample set for fine adjustment, and obtaining the target pavement leakage disease detector after preset iteration times are finished.
Preferably, the collecting of the road surface leakage disease images is performed by intercepting N images containing the road surface leakage disease samples from videos shot by the sweeper at each road section, and the image mode is RGB.
The invention has the following beneficial effects:
1. the yolov3 algorithm is high in accuracy and efficiency of identifying the road surface leakage diseases, and real-time target detection and identification of videos or images shot by the garbage sweeper can be achieved by using specific low-power-consumption edge GPU computing equipment. The recognition result provided by the algorithm can lay a good foundation for the pavement intelligent evaluation and scoring in the later period.
2. The iterative training method adopted by the invention has good generalization capability and can improve the identification precision of the unseen road scene and the unseen road leakage disease.
Drawings
The present invention will be further described and illustrated with reference to the following drawings.
Fig. 1 is a flow chart of an automatic road surface leakage disease detection method based on yolov 3.
Detailed Description
The technical solution of the present invention will be more clearly and completely explained by the description of the preferred embodiments of the present invention with reference to the accompanying drawings.
Examples
As shown in fig. 1, the method for automatically detecting the road surface leakage disease based on yolov3 provided by the invention comprises the following steps:
s1, collecting road surface leakage disease images to form a source sample set, and preprocessing the source sample set to obtain a training sample set;
in this step, the road surface leakage disease image is in an RGB mode.
In this step, the pre-processing includes horizontally flipping the image of the source sample set; randomly performing translation transformation, rotation transformation and color transformation on the image; and (5) carrying out image normalization processing.
In the step, the occurrence frequency of various broadcast leakage diseases in the source sample set is counted and sequenced, and the first N categories with the highest occurrence frequency are selected as N broadcast leakage diseases to be detected; and labeling each sample in the training sample set to obtain a sample label corresponding to the training sample set.
In this embodiment, a garbage sweeper running along a road surface shoots a video through a camera mounted on a sweeper body, 1000 original images containing road surface leakage disease samples are collected from the video as a source sample set, the pixel size of the original images is 4096 × 2160, small images covering the area of a road surface main body are intercepted and are uniformly zoomed in to 800 × 800, and an original data map is obtained.
Counting each category, selecting seven categories which affect the road appearance and have high occurrence frequency as categories of the pavement leakage diseases to be identified, wherein the categories are as follows: beverage bottles, plastic bags, boxes, paper sheets, wood chips, leaves, and crushed stones.
To increase the robustness of the detector, the scaled image is subjected to random horizontal flipping, translation transformation (transformation range between negative 5 pixels to positive 5 pixels and inclusive of negative 5 pixels and positive 5 pixels) and rotation transformation (transformation range between negative 30 degrees to positive 30 degrees and inclusive of negative 30 degrees and positive 30 degrees). And then, dividing the pixel value of the image by 255, and normalizing the value range of the pixel between 0 and 1 and including 0 and 1 to obtain a training sample set. And simultaneously, labeling the samples appearing in the training sample set to obtain a sample label.
And S2, training a yolov3 target detection network by utilizing a training sample set to obtain an initial pavement leakage disease detector.
In this step, there is further included the substeps of,
s201, training a yolov3 target detection network by adopting a stochastic gradient descent algorithm, setting the maximum iteration times, calculating a network error by adopting a minimum batch mode (mini-batch) in each iteration, and updating parameters in the yolov3 target detection network by using the network error. Wherein, the network error is the error between the classification label and the sample label output by the yolov3 target detection network.
And S202, terminating training when a preset maximum iteration number is reached or the error rate on the verification set is not reduced any more, and obtaining the initial pavement leakage disease detector.
In the present embodiment, the maximum number of iterations is set to 1500, the learning rate is set to 0.003, and 100 samples are input per iteration.
And S3, deploying an initial pavement leakage disease detector on the target road section to detect the pavement leakage diseases, and recording the detection result.
In this step, there is further included the substeps of,
s301, collecting the road surface leakage disease image again to serve as an original detection image;
s302, preprocessing an original detection image to obtain a detection image, and forming a detection sample set;
s303, extracting the characteristics of the detected image by using a backbone network in the initial leakage disease detector, and extracting all levels of characteristics of the network to form a characteristic pyramid;
s304, carrying out feature classification, wherein the feature classification is to use a classification network in the initial leakage disease detector to carry out feature classification on features, setting a threshold value, judging that the image contains the leakage disease to be detected if the output value is greater than the threshold value, and otherwise, judging that the image contains the background;
s305, drawing a leakage disease detection result in the detection image, wherein the detection result is the coordinate and confidence of the target output by the initial pavement leakage disease detector, and outputting the result to finish the leakage disease detection of the current image;
s306, storing the original detection image and the detection result.
S4, clustering and manually intervening the detection results and then labeling to obtain a target sample set, and carrying out online fine adjustment on the initial pavement leakage disease detector by using the target sample set, so that the accuracy of the target road section is improved, and the target pavement leakage disease detector is obtained.
In this step, there is further included the substeps of,
s401, setting a confidence threshold, traversing the confidence, directly adding the samples with the confidence greater than the confidence threshold into a training sample set, and adding the samples with the confidence less than the confidence threshold into a post-processing sample set;
s402, re-labeling the targets in the post-processing sample set, and adding the targets into the training sample set to obtain a target sample set;
and S403, repeating S201 and S202 on the initial pavement leakage disease detector by using the target sample set for fine adjustment, and obtaining the target pavement leakage disease detector after preset iteration times are finished.
The above detailed description merely describes preferred embodiments of the present invention and does not limit the scope of the invention. Without departing from the spirit and scope of the present invention, it should be understood that various changes, substitutions and alterations can be made herein by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims and their equivalents. The scope of the invention is defined by the claims.
Claims (10)
1. A pavement leakage disease automatic detection method based on yolov3 is characterized by comprising the following steps:
s1, collecting road surface leakage disease images to form a source sample set, and preprocessing the source sample set to obtain a training sample set;
s2, training a yolov3 target detection network by using the training sample set to obtain an initial pavement leakage disease detector;
s3, deploying the initial pavement leakage disease detector on the target road section to detect the pavement leakage diseases, and recording the detection result;
s4, clustering and manually intervening the detection results and then labeling to obtain a target sample set, and carrying out online fine adjustment on the initial pavement leakage disease detector by using the target sample set to improve the accuracy of the target road section and obtain the target pavement leakage disease detector.
2. The pavement leakage disease automatic detection method based on yolov3 as claimed in claim 1, wherein the S1 further comprises:
counting and sequencing the occurrence frequency of various broadcast leakage diseases in the source sample set, and selecting the first N categories with the highest occurrence frequency as N broadcast leakage diseases to be detected;
and labeling each sample in the training sample set to obtain a sample label corresponding to the training sample set.
3. The automatic pavement leakage disease detection method based on yolov3 as claimed in claim 1, wherein the pretreatment comprises:
horizontally turning the image to obtain an image to be further processed;
randomly carrying out translation transformation, rotation transformation and color transformation on the image to be further processed to obtain an image to be normalized;
and carrying out normalization processing on the image to be normalized.
4. The pavement leakage disease automatic detection method based on yolov3 as claimed in claim 2, wherein the S2 further comprises:
s201, training the yolov3 target detection network by adopting a random gradient descent algorithm, setting the maximum iteration times, calculating a network error by adopting a minimum batch mode (mini-batch) in each iteration, and updating parameters in the yolov3 target detection network by using the network error;
s202, terminating training when the preset maximum iteration times are reached or the error rate on the verification set is not reduced any more, and obtaining the initial pavement leakage disease detector.
5. The automatic pavement leakage disease detection method based on yolov3 of claim 4, wherein the network error is an error between a classification label output by the yolov3 target detection network and the sample label.
6. The automatic pavement leakage disease detection method based on yolov3 of claim 1, wherein S3 comprises:
s301, collecting the road surface leakage disease image again to serve as an original detection image;
s302, preprocessing the original detection image to obtain a detection image, and forming a detection sample set;
s303, extracting the characteristics of the detected image by using the backbone network in the initial leakage disease detector, and extracting all levels of characteristics of the network to form a characteristic pyramid;
s304, carrying out feature classification on the features;
s305, drawing a detection result of the broadcast leakage disease in the detection image, outputting the result and finishing the broadcast leakage disease detection of the current image;
s306, saving the original detection image and the detection result.
7. The method for automatically detecting the road surface leakage disease of yolov3 as claimed in claim 6, wherein the feature classification is to classify the features by using a classification network in the initial leakage disease detector, a threshold value is set, if the output value is greater than the threshold value, the image is determined to contain the leakage disease to be detected, otherwise, the image is determined to be background.
8. The method of claim 6, wherein the detection result is coordinates and confidence of the target output by the initial pavement leakage disease detector.
9. The pavement leakage disease automatic detection method based on yolov3 of claim 2, wherein the S4 comprises:
s401, setting a confidence threshold, traversing the confidence, directly adding the sample with the confidence greater than the confidence threshold into the training sample set, and adding the sample with the confidence less than the confidence threshold into a post-processing sample set;
s402, re-labeling the targets in the post-processing sample set, and adding the targets into the training sample set to obtain the target sample set;
and S403, repeating S201 and S202 on the initial pavement leakage disease detector by using the target sample set for fine adjustment, and obtaining the target pavement leakage disease detector after preset iteration times are finished.
10. The automatic pavement leakage disease detection method based on yolov3 as claimed in claim 1 or 6, wherein the collecting pavement leakage disease images is obtained by cutting N images containing pavement leakage disease samples from videos shot by a sweeper of each road section, and the image mode is RGB.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010875105.7A CN112115982A (en) | 2020-08-27 | 2020-08-27 | Yoov 3-based automatic detection method for road surface leakage diseases |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010875105.7A CN112115982A (en) | 2020-08-27 | 2020-08-27 | Yoov 3-based automatic detection method for road surface leakage diseases |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112115982A true CN112115982A (en) | 2020-12-22 |
Family
ID=73804946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010875105.7A Pending CN112115982A (en) | 2020-08-27 | 2020-08-27 | Yoov 3-based automatic detection method for road surface leakage diseases |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112115982A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110263675A (en) * | 2019-06-03 | 2019-09-20 | 武汉联一合立技术有限公司 | Garbage target recognition system and recognition method of community security robot |
| CN110533640A (en) * | 2019-08-15 | 2019-12-03 | 北京交通大学 | Based on the track circuit disease discrimination method for improving YOLOv3 network model |
| CN111523536A (en) * | 2020-04-10 | 2020-08-11 | 南京智行信息科技有限公司 | Self-adaptive road surface projectile intelligent detection method based on fast RCNN |
| CN111524121A (en) * | 2020-04-22 | 2020-08-11 | 南京智行信息科技有限公司 | Road and bridge fault automatic detection method based on machine vision technology |
-
2020
- 2020-08-27 CN CN202010875105.7A patent/CN112115982A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110263675A (en) * | 2019-06-03 | 2019-09-20 | 武汉联一合立技术有限公司 | Garbage target recognition system and recognition method of community security robot |
| CN110533640A (en) * | 2019-08-15 | 2019-12-03 | 北京交通大学 | Based on the track circuit disease discrimination method for improving YOLOv3 network model |
| CN111523536A (en) * | 2020-04-10 | 2020-08-11 | 南京智行信息科技有限公司 | Self-adaptive road surface projectile intelligent detection method based on fast RCNN |
| CN111524121A (en) * | 2020-04-22 | 2020-08-11 | 南京智行信息科技有限公司 | Road and bridge fault automatic detection method based on machine vision technology |
Non-Patent Citations (1)
| Title |
|---|
| 金瑶: ""城市道路视频中小像素目标检测"", 《光电工程》, pages 1 - 8 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Qu et al. | Crack detection of concrete pavement with cross-entropy loss function and improved VGG16 network model | |
| CN111080620B (en) | Road disease detection method based on deep learning | |
| CN106778835B (en) | Remote sensing image airport target identification method fusing scene information and depth features | |
| Rachmadi et al. | Vehicle color recognition using convolutional neural network | |
| CN108304798B (en) | Street level order event video detection method based on deep learning and motion consistency | |
| US8655070B1 (en) | Tree detection form aerial imagery | |
| CN105046252B (en) | A kind of RMB prefix code recognition methods | |
| CN103258213B (en) | A kind of for the dynamic vehicle model recognizing method in intelligent transportation system | |
| CN106022232A (en) | License plate detection method based on deep learning | |
| CN107609512A (en) | A kind of video human face method for catching based on neutral net | |
| CN111523536A (en) | Self-adaptive road surface projectile intelligent detection method based on fast RCNN | |
| CN105260749B (en) | Real-time target detection method based on direction gradient binary pattern and soft cascade SVM | |
| CN107798335A (en) | A kind of automobile logo identification method for merging sliding window and Faster R CNN convolutional neural networks | |
| CN103279752B (en) | A kind of eye locating method based on improving Adaboost algorithm and Face geometric eigenvector | |
| CN106529532A (en) | License plate identification system based on integral feature channels and gray projection | |
| CN106778742B (en) | Car logo detection method based on Gabor filter background texture suppression | |
| CN114749342B (en) | Lithium battery pole piece coating defect identification method, device and medium | |
| CN102998316A (en) | Transparent liquid impurity detection system and detection method thereof | |
| CN113221881B (en) | Multi-level smart phone screen defect detection method | |
| CN109635634A (en) | A kind of pedestrian based on stochastic linear interpolation identifies data enhancement methods again | |
| CN110751619A (en) | Insulator defect detection method | |
| CN109448307A (en) | A kind of recognition methods of fire disaster target and device | |
| CN109343701A (en) | A kind of intelligent human-machine interaction method based on dynamic hand gesture recognition | |
| Zhao et al. | Research on detection method for the leakage of underwater pipeline by YOLOv3 | |
| CN111950357A (en) | Marine water surface garbage rapid identification method based on multi-feature YOLOV3 |
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 |