CN114782409A - Vehicle surface anomaly detection method based on side direction camera - Google Patents
Vehicle surface anomaly detection method based on side direction camera Download PDFInfo
- Publication number
- CN114782409A CN114782409A CN202210564080.8A CN202210564080A CN114782409A CN 114782409 A CN114782409 A CN 114782409A CN 202210564080 A CN202210564080 A CN 202210564080A CN 114782409 A CN114782409 A CN 114782409A
- Authority
- CN
- China
- Prior art keywords
- image
- vehicle
- convolutional
- network
- images
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 14
- 238000012549 training Methods 0.000 claims description 18
- 230000011218 segmentation Effects 0.000 claims description 17
- 238000000034 method Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 13
- 230000005856 abnormality Effects 0.000 claims description 10
- 238000013527 convolutional neural network Methods 0.000 claims description 8
- 238000003709 image segmentation Methods 0.000 claims description 8
- 230000000873 masking effect Effects 0.000 claims description 8
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000012795 verification Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 2
- 230000007547 defect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005498 polishing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06T2207/10016—Video; Image sequence
-
- 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
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to the technical field of vehicle surface detection, and relates to a vehicle surface anomaly detection method based on a side direction camera.
Description
Technical Field
The invention belongs to the technical field of vehicle surface detection, and particularly relates to a vehicle surface anomaly detection method based on a side direction camera.
Background
With the rapid development of technologies such as artificial intelligence and computer vision, smart traffic plays an important role in smart cities, the smart traffic covers the aspects of roads in the cities, one problem which is easy to ignore in the smart traffic is that the anomalies on the surfaces of vehicles, such as slight scratches, breakage and the like, are often ignored or difficultly found by human eyes, and if the anomalies are not processed for a long time, the use of the vehicles is affected, and currently, a vehicle surface flaw detection method and a vehicle surface flaw detection system are disclosed in, for example, CN 110596116A; the vehicle surface flaw detection method comprises the following steps: s1, uniformly polishing the surface to be measured of the vehicle, and then continuously collecting images of the surface to be measured of the vehicle by using a camera to obtain a sampling image; s2, carrying out image fusion on all the sampling images to obtain a spliced image; s3, processing the spliced image to display a defect area, introducing new detection equipment in addition to the camera, for example, detecting the whole vehicle surface after being polished, wherein the technologies can only detect large scratches, large concave-convex parts and the like on the vehicle surface, the detection is difficult for small defects and small scratches on the vehicle surface, and the introduction of the new detection equipment increases the deployment difficulty for field deployment. Therefore, it is necessary to design a novel method for detecting surface abnormality of a vehicle, which can accurately detect an abnormal area on the surface of the vehicle based on existing equipment.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a vehicle surface anomaly detection method based on a side direction camera.
In order to achieve the above purpose, the specific process of the invention for detecting the surface abnormality of the vehicle is as follows:
(1) constructing a vehicle segmentation data set: collecting video images of a roadside parking lot side direction camera, segmenting and labeling the vehicle surface in the images, classifying pixels in the images into a vehicle surface class and a background class to obtain segmented data sets, and then dividing the segmented data sets into a training set, a verification set and a test set according to actual needs;
(2) segmenting the vehicle surface image by utilizing a segmentation network: scaling pixels of the vehicle surface image to 1333 x 800, selecting an image segmentation network to segment the vehicle surface image, and setting energy loss of the vehicle surface image;
(3) masking and repairing the surface image of the vehicle: scaling the surface image of each vehicle obtained in the step (2) to 512 × 256 resolution, sequentially performing mask processing on each vehicle when two vehicles are divided in the image acquired by the fisheye camera, wherein the mask size is 64 × 32 pixels, the mask processing is sequentially performed from the upper left corner to the lower right corner, the background part is not processed, and the number of masks of each vehicle is 64 at most; selecting 64 images from the images after the mask processing, inputting the 64 images into a convolutional neural network together for repairing to obtain the images after the mask repairing;
(4) comparing the original image with the image after the mask is repaired: comparing the image obtained by the step (3) after the mask is repaired with the original image, judging the similarity of each 64 positions for image repair, judging that the position has no damage if the original surface image is similar to the repaired image, and judging that the position has the damage if the surface image is not similar to the repaired image;
(5) training the network to obtain a trained parameter model: making a data set by using the segmentation result in the step (2), selecting 1000 segmented images for zooming, sequentially masking each image from the upper left corner to the lower right corner to serve as an input training sample of the network, outputting the images without masking by the network, calculating loss errors by adopting absolute value loss, updating parameters through back propagation, performing training iteration of a complete training set for a set number of times (500 times), storing model parameters with the best result on a verification set to serve as parameters with the trained final model, and obtaining the trained image repairing network parameters;
(6) the result is obtained by reasoning: and (4) inputting the video image of the side position monitoring camera into the image repairing network trained in the step (5) to carry out forward reasoning to obtain a repaired vehicle surface image, and judging whether the shot vehicle has surface abnormality or not through the step (4).
Further, the image Segmentation network model selected in the step (2) is a Dynamic and Fast semantic Segmentation model (refer to SOLOV2: Dynamic and Fast Instance Segmentation (2020 NeuroIPS)), and the network structure and the Segmentation energy loss are both structure and loss functions adopted in the document.
Further, in the step (3), the convolutional neural network includes nine convolution modules, which are a convolution module 1, a convolution module 2, a convolution module 3, a convolution module 4, a convolution module 5, a convolution module 4, a convolution module 3, a convolution module 2, and a convolution module 1 in sequence from left to right, a convolution kernel of a first layer of convolution layer in the leftmost convolution module 1 is 192 × 3, 192 indicates that 64 color images are input to the convolutional neural network together as a group of images, 3 × 3 indicates that 3 × 3 convolution is performed in an image coordinate plane, the layer has 64 convolution kernels in total, a convolution kernel of a second layer of convolution layer in the convolution module 1 is 64 × 3, a size of the obtained feature layer is 64 × 512 × 256, 64 is a thickness of the feature layer, 512 × 256 is a dimension of the feature layer plane, and the number of the convolution kernels in the modules 2, 3, 4, and 5 is 128 in sequence, 256. 512 and 512.
Further, the similarity judgment in the step (4) adopts a structural similarity criterion (SSIM) for judgment, and the SSIM calculation process is as follows:
，
wherein the sums are respectively original imagesI g And repairing the imageI d And are respectively the original imageI g And repairing the imageI d Standard deviation of (a), covariance, c1And c2Is a constant number c1Taking 6.5025, c258.5225 is taken; the value range of SSIM is 0-1, the closer to 1, the image obtained by restoration is to the original real image, if the SSIM is more than or equal to 0.8, the original surface image is judged to be not abnormal, and if the SSIM is more than or equal to 0.8<And 0.8, judging that the original vehicle surface has damage, and giving an alarm.
Further, the picture size of the training sample input in the step (5) is
According to the batch size B, the batch size B is sequentially input into the network, and the input of the whole network is
。
Compared with the prior art, the invention has the beneficial effects that: training an image segmentation network by constructing a data set, constructing an image restoration data set for the segmented image, constructing an image restoration network model, performing restoration processing after mask processing on the original image, calculating the similarity between the original image and the restored image by a structural similarity criterion, considering that the image is abnormal if the similarity is smaller than a threshold, considering that no abnormality exists if the similarity is larger than or equal to the threshold, traversing from the upper left corner to the lower right corner, considering that the visible part of the vehicle is abnormal if all image blocks are similar to the original image after restoration, and considering that the surface of the vehicle is abnormal if abnormal blocks exist, and the method can quickly and effectively detect the abnormality of the surface of the vehicle.
Drawings
Fig. 1 is a diagram illustrating an image restoration network according to the present invention.
Fig. 2 is a schematic view of one of the masks of the vehicle surface of the present invention, in which 6 is a vehicle and 7 is a mask.
Fig. 3 is a schematic view of the working process of the present invention.
Detailed Description
The invention will be further described by way of examples in connection with the accompanying drawings without in any way limiting the scope of the invention.
Example (b):
the specific process of detecting the vehicle surface abnormality based on the side direction camera in the embodiment of the present invention is shown in fig. 3, and specifically includes the following steps:
(1) vehicle segmentation data set construction:
collecting video images of a roadside parking lot side direction camera, segmenting and labeling the vehicle surface in the images, and classifying pixels in the images into a vehicle surface class and a background class;
(2) segmenting the vehicle surface image by using an example segmentation network:
utilizing the vehicle segmentation data constructed in the step (1), selecting an image segmentation network to segment the surface area of the vehicle, and setting the energy loss of the image segmentation network, wherein the selected image segmentation network model is as follows: dynamic and Fast semantic Segmentation model (refer to SOLOV2: Dynamic and Fast Instance Segmentation (2020 NeuroIPS)), network structure and Segmentation energy loss are both structure and loss functions adopted in the document; because the resolution of the image collected by the original side phase camera is 1920 × 1080, the image is firstly zoomed to 1333 × 800 before being divided, and the size pixel of the output divided image is also 1333 × 800;
(3) carrying out mask processing on the vehicle surface image and repairing:
scaling the surface image of each vehicle obtained in the step (2) to 512 × 256 resolution, and for the fish-eye camera, if the image is divided into two vehicles, sequentially performing mask processing on each vehicle, wherein the mask size is 64 × 32 pixels, the mask processing is performed from the upper left corner to the lower right corner, and the background part does not need to be processed, so that the number of masks of each vehicle is 64 at most;
constructing an image restoration network as shown in fig. 1, inputting 64 images into a convolutional neural network together, unlike general image restoration, a general image restoration algorithm is to input one image with a mask, output as a mask-restored image, and our algorithm is to input 64 images with masks at a time, output as the images after mask restoration, in the leftmost convolution module 1 of the image restoration network, the convolution kernel of the first convolution layer is 192 × 3, 192 indicates that 64 color images are input to the convolutional neural network together as an image group, 3 × 3 indicates that 3 × 3 convolution is adopted in the image coordinate plane, the layer has 64 convolution kernels, the convolution of the second convolution layer in the convolution module is 64 x 3, the size of the obtained feature layer is 64 x 512 x 256, 64 is the thickness of the feature layer, and 512 x 256 is the dimension of the feature layer plane; the number of convolution kernels in the convolution modules 2, 3, 4 and 5 is 128, 256, 512 and 512 in sequence;
(4) comparing the original image with the restored image result to detect the surface abnormality of the vehicle:
comparing the repairing result obtained in the step (3) with the original image, judging the similarity of each of 64 positions for image repairing, if the original surface image is similar to the repaired image, judging that the position has no damage, if the surface image is not similar to the repaired image, judging that the position has the damage, and judging the similarity to be a structural similarity criterion (SSIM),
the SSIM calculation method is as follows:
，
wherein the sums are respectively original imagesI g And repairing the imageI d And are respectively the original imageI g And repairing the imageI d Standard deviation of (a), as covariance, c1And c2For constant, 6.5025 is taken as c1, and 58.5225 is taken as c 2; the value range of SSIM is 0-1, the closer to 1, the image obtained by repairing is to the original real image, if the SSIM is more than or equal to 0.8, the original surface image is judged to have no abnormity, and if the SSIM is more than or equal to 0.8<0.8, judging that the original vehicle surface has damage, and giving an alarm;
(5) training the network to obtain a trained parameter model:
making a data set by using the segmentation result in the step (2), selecting 1000 segmented images for zooming, sequentially masking each image from the upper left corner to the lower right corner, and taking the images as input training samples of the network, wherein the network output is images without masking, and the size of the input images is
Input into the network in turn according to the batch size (B), input of the entire network
Calculating loss errors by adopting absolute loss, updating parameters through back propagation, saving model parameters with the best results on a verification set after 500 times of training iterations of a complete training set, and taking the model parameters as final model trained parameters to obtain trained image repairing network parameters;
(6) the result is obtained by reasoning:
and (4) inputting the video image of the side position monitoring camera into the network trained in the step (5) to carry out forward reasoning to obtain a repaired vehicle surface image, and judging whether the shot vehicle has surface abnormality or not through the step (4).
Techniques not described in detail herein are common or prior art in the art.
It is noted that the present embodiment is intended to aid in further understanding of the present invention, but those skilled in the art will understand that: various substitutions and modifications are possible without departing from the spirit and scope of the invention and appended claims. Therefore, the invention should not be limited to the embodiments disclosed, but the scope of the invention is defined by the appended claims.
Claims (5)
1. A vehicle surface anomaly detection method based on a side direction camera is characterized by comprising the following specific processes:
(1) constructing a vehicle segmentation data set: collecting video images of a roadside parking lot side direction camera, segmenting and labeling the vehicle surface in the images, classifying pixels in the images into a vehicle surface class and a background class to obtain segmented data sets, and then dividing the segmented data sets into a training set, a verification set and a test set according to actual needs;
(2) segmenting the vehicle surface image by utilizing a segmentation network: scaling pixels of the vehicle surface image to 1333 × 800, selecting an image segmentation network to segment the vehicle surface image, and setting energy loss of the vehicle surface image;
(3) masking and repairing the surface image of the vehicle: scaling the surface image of each vehicle obtained in the step (2) to 512 × 256 resolution, sequentially performing mask processing on each vehicle when two vehicles are divided in the image acquired by the fish-eye camera, wherein the mask size is 64 × 32 pixels, performing mask processing from the upper left corner to the lower right corner, processing no background part, and making 64 masks of each vehicle at most; selecting 64 images from the images after the mask processing, inputting the images into a convolutional neural network together for repairing to obtain an image after the mask repairing;
(4) comparing the original image with the image after the mask is repaired: comparing the image obtained by the step (3) after the mask is repaired with the original image, judging the similarity of each of 64 positions for image repair, judging that the position has no damage if the original surface image is similar to the repaired image, and judging that the position has the damage if the surface image is not similar to the repaired image;
(5) training the network to obtain a trained parameter model: making a data set by using the segmentation result in the step (2), selecting 1000 segmented images for zooming, sequentially masking each image from the upper left corner to the lower right corner to serve as an input training sample of the network, outputting the images without masking by the network, calculating loss errors by adopting absolute value loss, updating parameters through back propagation, storing model parameters with the best results on a verification set after training iteration of the complete training set for a set number of times, and taking the model parameters as parameters with the good results of the final model training to obtain the trained image repairing network parameters;
(6) the result is obtained by reasoning: and (4) inputting the video image of the side position monitoring camera into the image repairing network trained in the step (5) to carry out forward reasoning to obtain a repaired vehicle surface image, and judging whether the shot vehicle has surface abnormality or not through the step (4).
2. The method for detecting anomalies on a vehicle surface based on a side-oriented camera of claim 1, characterized in that the image segmentation network model selected in step (2) is a dynamic and fast semantic segmentation model.
3. The method according to claim 2, wherein the convolutional neural network in step (3) comprises nine convolutional modules, namely a convolutional module 1, a convolutional module 2, a convolutional module 3, a convolutional module 4, a convolutional module 5, a convolutional module 4, a convolutional module 3, a convolutional module 2 and a convolutional module 1 from left to right, the convolutional kernel of the first convolutional layer in the leftmost convolutional module 1 is 192 x 3, 192 represents that 64 color images are input to the convolutional neural network together as an image group, 3 x 3 represents that 3 x 3 convolution is adopted in the image coordinate plane, the layer has 64 convolutional kernels in total, the convolution of the second convolutional layer in the convolutional module 1 is 64 x 3, the size of the obtained feature layer is 64 x 512, 256 is the thickness of the feature layer, and 512 is the dimension of the feature layer plane, in the convolution modules 2, 3, 4 and 5, the number of convolution kernels is 128, 256, 512 and 512 in sequence.
4. The method for detecting the vehicle surface abnormality based on the side orientation camera according to claim 3, wherein the similarity judgment in the step (4) is performed by adopting a structural similarity criterion SSIM, and the SSIM calculation process is as follows:
wherein the sums are respectively original imagesI g And repairing the imageI d And are respectively the original imageI g And repairing the imageI d Standard deviation of (a), as covariance, c1And c2Is a constant, c1Taking 6.5025, c258.5225 is taken; the value range of SSIM is 0-1, the closer to 1, the image obtained by restoration is to the original real image, if the SSIM is more than or equal to 0.8, the original surface image is judged to be not abnormal, and if the SSIM is more than or equal to 0.8<And 0.8, judging that the original vehicle surface has damage, and giving an alarm.
5. The method of claim 4, wherein the training sample picture inputted in the step (5) has a size of
According to the batch size B, the batch size B is sequentially input into the network, and the input of the whole network is
。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210564080.8A CN114782409B (en) | 2022-05-23 | 2022-05-23 | Vehicle surface abnormality detection method based on side direction camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210564080.8A CN114782409B (en) | 2022-05-23 | 2022-05-23 | Vehicle surface abnormality detection method based on side direction camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114782409A true CN114782409A (en) | 2022-07-22 |
| CN114782409B CN114782409B (en) | 2024-03-29 |
Family
ID=82408793
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210564080.8A Active CN114782409B (en) | 2022-05-23 | 2022-05-23 | Vehicle surface abnormality detection method based on side direction camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114782409B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2020103901A4 (en) * | 2020-12-04 | 2021-02-11 | Chongqing Normal University | Image Semantic Segmentation Method Based on Deep Full Convolutional Network and Conditional Random Field |
| CN113095277A (en) * | 2021-04-27 | 2021-07-09 | 东南大学 | Unmanned aerial vehicle aerial photography vehicle detection method based on target space distribution characteristics |
| CN113658115A (en) * | 2021-07-30 | 2021-11-16 | 华南理工大学 | Image anomaly detection method for generating countermeasure network based on deep convolution |
-
2022
- 2022-05-23 CN CN202210564080.8A patent/CN114782409B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2020103901A4 (en) * | 2020-12-04 | 2021-02-11 | Chongqing Normal University | Image Semantic Segmentation Method Based on Deep Full Convolutional Network and Conditional Random Field |
| CN113095277A (en) * | 2021-04-27 | 2021-07-09 | 东南大学 | Unmanned aerial vehicle aerial photography vehicle detection method based on target space distribution characteristics |
| CN113658115A (en) * | 2021-07-30 | 2021-11-16 | 华南理工大学 | Image anomaly detection method for generating countermeasure network based on deep convolution |
Non-Patent Citations (1)
| Title |
|---|
| 阮航;王立春;: "基于特征图的车辆检测和分类", 计算机技术与发展, no. 11, 30 June 2018 (2018-06-30), pages 39 - 43 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114782409B (en) | 2024-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106875381B (en) | Mobile phone shell defect detection method based on deep learning | |
| Hou et al. | Inspection of surface defects on stay cables using a robot and transfer learning | |
| Dong et al. | A deep-learning-based multiple defect detection method for tunnel lining damages | |
| Li et al. | Automatic crack detection and measurement of concrete structure using convolutional encoder-decoder network | |
| CN111223093A (en) | AOI defect detection method | |
| CN111080620A (en) | Road disease detection method based on deep learning | |
| CN109671071B (en) | Underground pipeline defect positioning and grade judging method based on deep learning | |
| CN114549997B (en) | X-ray image defect detection method and device based on regional feature extraction | |
| CN110838112A (en) | Insulator defect detection method based on Hough transform and YOLOv3 network | |
| WO2021238420A1 (en) | Image defogging method, terminal, and computer storage medium | |
| CN110555831B (en) | Deep learning-based drainage pipeline defect segmentation method | |
| CN114898352A (en) | Method for simultaneously realizing image defogging and license plate detection | |
| Ma et al. | Automatic defogging, deblurring, and real-time segmentation system for sewer pipeline defects | |
| CN107346547A (en) | Real-time foreground extracting method and device based on monocular platform | |
| CN116612106A (en) | Method for detecting surface defects of optical element based on YOLOX algorithm | |
| CN115830004A (en) | Surface defect detection method, device, computer equipment and storage medium | |
| CN114170144A (en) | Power transmission line pin defect detection method, equipment and medium | |
| CN113240623A (en) | Pavement disease detection method and device | |
| Peng et al. | Non-uniform illumination image enhancement for surface damage detection of wind turbine blades | |
| CN116205876A (en) | Unsupervised notebook appearance defect detection method based on multi-scale standardized flow | |
| CN114972202A (en) | Ki67 pathological cell rapid detection and counting method based on lightweight neural network | |
| Shit et al. | An encoder‐decoder based CNN architecture using end to end dehaze and detection network for proper image visualization and detection | |
| CN113870204A (en) | Method and device for detecting abnormality of optical glass | |
| CN111612803B (en) | Vehicle image semantic segmentation method based on image definition | |
| CN111325724B (en) | Tunnel crack region detection method and device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |