CN111428563B - Image identification method for automobile full-liquid crystal instrument - Google Patents
Image identification method for automobile full-liquid crystal instrumentDownload PDFInfo
- Publication number
- CN111428563B CN111428563BCN202010114458.5ACN202010114458ACN111428563BCN 111428563 BCN111428563 BCN 111428563BCN 202010114458 ACN202010114458 ACN 202010114458ACN 111428563 BCN111428563 BCN 111428563B
- Authority
- CN
- China
- Prior art keywords
- image
- liquid crystal
- crystal instrument
- full liquid
- point
- 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.)
- Active
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/30—Scenes; Scene-specific elements in albums, collections or shared content, e.g. social network photos or video
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/757—Matching configurations of points or features
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/02—Recognising information on displays, dials, clocks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
Translated fromChinese
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to an image recognition method for an automobile full-liquid-crystal instrument.
Background
With the development of electronic information technology, 5G and other communication technologies, the full liquid crystal instrument is widely applied and popularized in automobiles, image information contained in the full liquid crystal instrument is very rich, the detection technology of the display function of the full liquid crystal instrument mainly adopts manual visual inspection, and the detection speed and the detection efficiency are greatly influenced by human subjective factors.
Disclosure of Invention
The invention aims to solve the technical problem of providing an image identification method for an automobile full-liquid crystal instrument. By improving the scale space of the original SIFT algorithm, the generation of the key point descriptors, the matching of the feature points and the like, the efficiency and the real-time performance of recognition and matching are improved, and the method can be applied to the automatic test of the display function of the automobile full-liquid crystal instrument.
The invention is realized in this way, a method for identifying the image of the full liquid crystal instrument of the automobile comprises the following steps:
s1: collecting images of the full liquid crystal instrument in different functional states and creating a standard characteristic model library of the full liquid crystal instrument image;
s2: adopting a camera device to acquire images of the full liquid crystal instrument to be detected in different display function states under the condition that a dark box has no light source;
s3: carrying out enhancement processing on the acquired full liquid crystal instrument image, wherein the enhancement processing comprises image smoothing processing and image sharpening processing;
s4: carrying out feature point matching on the full liquid crystal instrument image to be detected and a standard image in a feature model library by utilizing an improved SIFT algorithm;
s5: and obtaining the recognition result according to the set matching rate threshold value.
Wherein: the specific steps of carrying out feature point matching on the full liquid crystal instrument image to be detected and the standard image in the feature model library by using the improved SIFT algorithm are as follows:
s41, simplifying the scale space, and simplifying the 5-layer number and the 5-layer number of the scale space in the original algorithm into 4 layers;
s42, detecting the extreme point of the scale space, comparing the sampling point with all the points adjacent to the space to determine the extreme point;
s43, locating extreme points, and eliminating unstable and low-contrast extreme points;
s44, determining the principal direction of the key point, taking the extreme point after positioning as the feature point, calculating the gradient and direction of each feature point, and then counting in the gradient direction of the pixel point by a histogram method, wherein the peak value of the histogram is used for representing the principal direction of the key point;
s45: simplifying the key point descriptors, namely simplifying the key point descriptors in the original SIFT algorithm from 128-dimensional vectors to 24-dimensional vectors;
s46: and matching the characteristic points, namely matching the characteristic points by adopting a Randac algorithm.
Further, the camera device in step S2 takes a picture of the full liquid crystal instrument to be detected under the condition of dark installation and no light source, the sampled picture is uploaded to an upper computer to be processed, the camera is a CCD camera, and the pixel size is 2560 × 1920.
Further, the step S3 includes the specific steps of the full liquid crystal instrument image enhancement processing:
s31: and (3) image smoothing processing: smoothing the image by adopting a median filtering method: the conversion formula is as follows:
g(x,y)＝median{f(x-k,y-l),(k,l)∈W}
wherein f (x, y) is an original image, g (x, y) is a filtered output image, W is a median filtering template, and l and k are boundary values in x and y directions respectively;
s32: image sharpening processing: adopting a Laplace operator to sharpen the image: the sharpening formula is as follows:
wherein f (x, y) is an original image, g (x, y) is a sharpened output image, and c is a laplacian sharpened template.
Further, S43, locating the extreme points, and eliminating the unstable and low-contrast extreme points includes: fitting function formula in scale space by using Gaussian difference function:
where X ═ X, y, σ)TThe derivation can obtain the value of the corresponding extreme point equation
If it is
The extreme point is eliminated and the extreme point is eliminated,
represents the offset of the extreme point, and D (X) is a Gaussian difference function value.
Further, S45: the method for simplifying the key point descriptor includes the following steps:
changing a rectangular region in the original SIFT algorithm into a concentric region, taking a key point as a circle center, taking a diameter of 8, and dividing concentric circles with radiuses of 1, 2, 3 and 4 respectively;
respectively calculating gradients of four concentric circles in 6 directions of 0 degrees, 60 degrees, 120 degrees, 180 degrees, 240 degrees and 300 degrees, wherein each concentric circle forms an 8-dimensional vector;
a simplified 4 x 6-24 keypoint descriptor was obtained.
Further, step S46 includes:
(1) setting the acquired key point descriptor sample set as N, determining the minimum number of samples required by the model parameter as N, ensuring that the number of the samples in the N is greater than the value of N, and randomly selecting a subset S of the N containing r samples and an initialization model W in the sample set N;
(2) taking a residual set N/S, taking a sample set with the error of W being less than a certain threshold value t, and forming an inner point set S together with S*；
(3) If S is*N or more, representing that the correct model parameter is generated according to the inner point set S*Recalculating a new model W, selecting a new subset S, and performing k iterations on the process;
(4) after k iterations, when the maximum consistent set Y is found, the interior and exterior points in the dataset are judged according to Y.
Compared with the prior art, the invention has the beneficial effects that:
the method can reduce the matching time and improve the matching efficiency and the matching real-time performance of the SIFT algorithm. The invention is used for image recognition of the full liquid crystal instrument and can realize automatic testing of the full liquid crystal instrument in different functional states.
Drawings
FIG. 1 is a flow chart of a method of the present invention;
FIG. 2 is a flow chart of the improved SIFT algorithm of the present invention;
FIG. 3 is a schematic diagram of a simplified key point descriptor of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, the invention provides an image recognition method for an automobile full liquid crystal instrument based on an improved SIFT algorithm, which specifically comprises the following steps:
step 1: collecting images of the full liquid crystal instrument in different functional states and creating a standard characteristic model library of the full liquid crystal instrument image;
step 2: adopting a camera device to acquire images of the full liquid crystal instrument to be detected in different display function states under the condition that a dark box has no light source; the camera device shoots the full liquid crystal instrument to be detected under the condition of dark installation without a light source, a sampling photo is uploaded to an upper computer to be processed, a CCD camera is selected as the camera, and the pixel size is 2560 multiplied by 1920.
And step 3: the image enhancement processing of the full liquid crystal instrument in the upper computer specifically comprises the following steps:
step 31: and (3) smoothing the image, namely smoothing the image by adopting a median filtering method, wherein a conversion formula is as follows:
g(x,y)＝median{f(x-k,y-l),(k,l)∈W}
wherein f (x, y) is an original image, g (x, y) is a filtered output image, W is a median filtering template, and l and k are boundary values in x and y directions respectively.
Step 32: and (5) image sharpening processing. Adopting a Laplace operator to sharpen the image, wherein the sharpening formula is as follows:
wherein f (x, y) is an original image, g (x, y) is a sharpened output image, and c is a laplacian sharpened template.
And 4, step 4: the method comprises the following steps of carrying out feature point matching on a full liquid crystal instrument image to be detected and a standard image in a feature model library by utilizing an improved SIFT algorithm, and specifically comprises the following steps:
and 41, simplifying the scale space, and simplifying the 5-layer number and the 5-layer number of the scale space in the original algorithm into 4 layers.
And 42, detecting the extreme point of the scale space, and comparing the sampling point with all the points adjacent to the space to determine the extreme point.
And 43, positioning extreme points and removing unstable and low-contrast extreme points. Fitting function formula in scale space by using Gaussian difference function:
where X ═ X, y, σ)TThe derivation can obtain the value of the corresponding extreme point equation
If it is
The extreme point is eliminated and the extreme point is eliminated,
represents the offset of the extreme point, and D (X) is a Gaussian difference function value.
And 44, determining the main direction of the key point, wherein the positioned extreme point is a characteristic point. After the gradient and the direction of each feature point are calculated, statistics is carried out on the gradient direction of the pixel points by a histogram method, and the peak value of the histogram is used for representing the main direction of the key point.
Step 45: the keypoint descriptor is simplified. The key point descriptors in the original algorithm are reduced from 128-dimensional vectors to 24-dimensional vectors. Referring to fig. 3, the simplified method is as follows:
(1) a rectangular region in the original SIFT algorithm is changed into a concentric circle region, a key point is used as a circle center, the diameter is 8, and concentric circles with the radiuses of 1, 2, 3 and 4 are respectively divided.
(2) Gradients of four concentric circles in 6 directions of 0 degrees, 60 degrees, 120 degrees, 180 degrees, 240 degrees and 300 degrees are respectively calculated, and each concentric circle forms an 8-dimensional vector.
(3) A simplified 4 x 6-24 keypoint descriptor can be obtained.
Step 46: and matching the characteristic points. And (4) carrying out feature point matching by adopting a Ranpac algorithm to replace an Euclidean distance method of the original algorithm. The method comprises the following steps: the Ransac algorithm steps are as follows:
(1) setting the acquired key point descriptor sample set as N, determining the minimum number of samples required by the model parameter as N, ensuring that the number of the samples in the N is greater than the value of N, and randomly selecting a subset S of the N containing r samples and an initialization model W in the sample set N;
(2) taking a residual set N/S, taking a sample set with the error of W being less than a certain threshold value t, and forming an inner point set S together with S*；
(3) If S is*N or more, representing that the correct model parameter is generated according to the inner point set S*Recalculating a new model W, selecting a new subset S, and performing k iterations on the process;
(4) after k iterations, when the maximum consistent set Y is found, the interior and exterior points in the dataset are judged according to Y.
And 5: the recognition result is obtained according to a matching rate threshold, which is generally set to 90%.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (4)
1. An image identification method for an automobile full liquid crystal instrument is characterized by comprising the following steps:
s1: collecting images of the full liquid crystal instrument in different functional states and creating a standard characteristic model library of the full liquid crystal instrument image;
s2: adopting a camera device to acquire images of the full liquid crystal instrument to be detected in different display function states under the condition that a dark box has no light source;
s3: carrying out enhancement processing on the acquired full liquid crystal instrument image, wherein the enhancement processing comprises image smoothing processing and image sharpening processing;
s4: carrying out feature point matching on the full liquid crystal instrument image to be detected and a standard image in a feature model library by utilizing an improved SIFT algorithm;
s5: obtaining an identification result according to a set matching rate threshold;
wherein: the specific steps of carrying out feature point matching on the full liquid crystal instrument image to be detected and the standard image in the feature model library by using the improved SIFT algorithm are as follows:
s41, simplifying the scale space, and simplifying the 5-layer number and the 5-layer number of the scale space in the original algorithm into 4 layers;
s42, detecting the extreme point of the scale space, comparing the sampling point with all the points adjacent to the space to determine the extreme point;
s43, locating extreme points, and eliminating unstable and low-contrast extreme points;
s44, determining the principal direction of the key point, taking the extreme point after positioning as the feature point, calculating the gradient and direction of each feature point, and then counting in the gradient direction of the pixel point by a histogram method, wherein the peak value of the histogram is used for representing the principal direction of the key point;
s45: simplifying the key point descriptors, namely simplifying the key point descriptors in the original SIFT algorithm from 128-dimensional vectors to 24-dimensional vectors;
s46: matching the characteristic points, namely matching the characteristic points by adopting a Randac algorithm;
wherein, S45: the method for simplifying the key point descriptor includes the following steps:
changing a rectangular region in the original SIFT algorithm into a concentric region, taking a key point as a circle center, taking a diameter of 8, and dividing concentric circles with radiuses of 1, 2, 3 and 4 respectively;
respectively calculating gradients of four concentric circles in 6 directions of 0 degrees, 60 degrees, 120 degrees, 180 degrees, 240 degrees and 300 degrees, wherein each concentric circle forms an 8-dimensional vector;
obtaining simplified 4 x 6 ═ 24 key point descriptors;
step S46 includes:
(1) setting the acquired key point description sub-sample set as N, determining the minimum number of samples required by the model parameters as N, ensuring that the number of samples in N is greater than the value of N, and randomly selecting a subset S of N containing r samples and an initialization model W in the sample set N;
(2) taking a residual set N/S, taking a sample set with the error of W being less than a certain threshold value t, and forming an inner point set S together with S*；
(3) If S is*N or more, representing that the correct model parameter is generated according to the inner point set S*Recalculating a new model W, selecting a new subset S, and performing k iterations on the process;
(4) after k iterations, when the maximum consistent set Y is found, the interior and exterior points in the dataset are judged according to Y.
2. The method of claim 1, wherein the camera of step S2 is used to take a picture of the full liquid crystal instrument to be tested under dark-mounted and no-light conditions, and the sampled picture is uploaded to an upper computer to be processed, wherein the camera is a CCD camera with a pixel size of 2560 × 1920.
3. The method as claimed in claim 1, wherein the step S3 full liquid crystal instrument image enhancement processing comprises the following specific steps:
s31: and (3) image smoothing processing: smoothing the image by adopting a median filtering method: the conversion formula is as follows:
g(x,y)＝median{f(x-k,y-l),(k,l)∈W}
wherein f (x, y) is an original image, g (x, y) is a filtered output image, W is a median filtering template, and l and k are boundary values in x and y directions respectively;
s32: image sharpening processing: adopting a Laplace operator to sharpen the image: the sharpening formula is as follows:
wherein f (x, y) is an original image, g (x, y) is a sharpened output image, and c is a laplacian sharpened template.
4. The method of claim 1, wherein S43 locating the extreme points, and wherein culling the unstable and low contrast extreme points comprises: fitting function formula in scale space by using Gaussian difference function:
where X ═ X, y, σ)TThe derivation can obtain the value of the corresponding extreme point equation
If it is
The extreme point is eliminated and the extreme point is eliminated,
represents the offset of the extreme point, and D (X) is a Gaussian difference function value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010114458.5ACN111428563B (en) | 2020-02-25 | 2020-02-25 | Image identification method for automobile full-liquid crystal instrument |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010114458.5ACN111428563B (en) | 2020-02-25 | 2020-02-25 | Image identification method for automobile full-liquid crystal instrument |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111428563A CN111428563A (en) | 2020-07-17 |
| CN111428563Btrue CN111428563B (en) | 2022-04-01 |
Family
ID=71547226
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010114458.5AActiveCN111428563B (en) | 2020-02-25 | 2020-02-25 | Image identification method for automobile full-liquid crystal instrument |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111428563B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113110909A (en)* | 2021-04-20 | 2021-07-13 | 肇庆小鹏汽车有限公司 | Vehicle instrument testing method and device |
| CN114323303B (en)* | 2021-12-31 | 2023-08-29 | 深圳技术大学 | Body temperature measuring method, device, infrared thermometer and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722731A (en)* | 2012-05-28 | 2012-10-10 | 南京航空航天大学 | Efficient image matching method based on improved scale invariant feature transform (SIFT) algorithm |
| CN103927507A (en)* | 2013-01-12 | 2014-07-16 | 山东鲁能智能技术有限公司 | Improved multi-instrument reading identification method of transformer station inspection robot |
| CN105426929A (en)* | 2014-09-19 | 2016-03-23 | 佳能株式会社 | Object shape alignment device, object processing device and methods thereof |
| CN109063717A (en)* | 2018-07-30 | 2018-12-21 | 安徽慧视金瞳科技有限公司 | A kind of acquisition instrument center point method |
| CN109542041A (en)* | 2019-01-07 | 2019-03-29 | 吉林大学 | A kind of automation equipment based on the detection of automobile double screen |
| CN110044405A (en)* | 2019-05-16 | 2019-07-23 | 吉林大学 | A kind of automobile instrument automatic detection device and method based on machine vision |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012222674A (en)* | 2011-04-12 | 2012-11-12 | Sony Corp | Image processing apparatus, image processing method, and program |
| CN106529424B (en)* | 2016-10-20 | 2019-01-04 | 中山大学 | A kind of logo detection recognition method and system based on selective search algorithm |
- 2020
- 2020-02-25CNCN202010114458.5Apatent/CN111428563B/enactiveActive
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722731A (en)* | 2012-05-28 | 2012-10-10 | 南京航空航天大学 | Efficient image matching method based on improved scale invariant feature transform (SIFT) algorithm |
| CN103927507A (en)* | 2013-01-12 | 2014-07-16 | 山东鲁能智能技术有限公司 | Improved multi-instrument reading identification method of transformer station inspection robot |
| CN105426929A (en)* | 2014-09-19 | 2016-03-23 | 佳能株式会社 | Object shape alignment device, object processing device and methods thereof |
| CN109063717A (en)* | 2018-07-30 | 2018-12-21 | 安徽慧视金瞳科技有限公司 | A kind of acquisition instrument center point method |
| CN109542041A (en)* | 2019-01-07 | 2019-03-29 | 吉林大学 | A kind of automation equipment based on the detection of automobile double screen |
| CN110044405A (en)* | 2019-05-16 | 2019-07-23 | 吉林大学 | A kind of automobile instrument automatic detection device and method based on machine vision |
Non-Patent Citations (5)
| Title |
|---|
| "Vision based feature diagnosis for automobile instrument cluster using machine learning";M.Deepan Raj 等;《2017 Fourth International Conference on Signal Processing,Communication and Networking (ICSCN)》;20171030;第1-3页* |
| "图像识别技术在汽车仪表板检测中的应用与研究";张娇;《中国优秀硕士学位论文全文数据库(信息科技辑)》;20151215(第2015年第12期);第I138-699页正文第14-36页* |
| "基于SIFT的图像匹配方法的研究与改进";李宁;《万方数据(学位)》;20141013;第7-26页* |
| "基于快速SIFT特征提取的模板匹配算法";李忠海 等;《计算机工程》;20111220;第37卷(第24期);第223-224页* |
| "基于机器视觉的圆形指针式仪表自动读数识别关键技术研究";韩绍超;《中国优秀硕士学位论文全文数据库(信息科技辑)》;20190215(第2019年第02期);第I138-1295页正文第10-35页* |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111428563A (en) | 2020-07-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lopez-Molina et al. | Multiscale edge detection based on Gaussian smoothing and edge tracking | |
| CN108564092A (en) | Sunflower disease recognition method based on SIFT feature extraction algorithm | |
| CN111881923B (en) | Bill element extraction method based on feature matching | |
| CN111428563B (en) | Image identification method for automobile full-liquid crystal instrument | |
| CN117173437B (en) | Multimodal remote sensing image hybrid matching method and system | |
| CN105930852A (en) | Method for identifying bubble image | |
| CN107766864B (en) | Method and device for extracting features and method and device for object recognition | |
| CN111553422A (en) | Automatic identification and recovery method and system for surgical instruments | |
| CN113129260A (en) | Automatic detection method and device for internal defects of lithium battery cell | |
| CN119205786B (en) | Underwater pipeline detection method, device, equipment and storage medium | |
| CN112836726B (en) | Pointer instrument indication reading method and device based on video information | |
| CN110222661A (en) | It is a kind of for motion estimate and the feature extracting method of tracking | |
| CN115620132A (en) | An Unsupervised Method for Contrastive Learning Ice Lake Extraction | |
| CN106157240B (en) | Remote sensing image super-resolution method based on dictionary learning | |
| CN116433978A (en) | Automatic generation and automatic labeling method and device for high-quality flaw image | |
| CN115689929A (en) | Aperture Measurement Method and Device Based on Distortion Compensation and Adaptive Mean Fuzzy | |
| CN105761207B (en) | Image Super-resolution Reconstruction method based on the insertion of maximum linear block neighborhood | |
| CN118392815B (en) | Microplastic identification method | |
| CN106709516B (en) | A Blurred Image Detection Method Based on Naive Bayesian Method | |
| CN117876270B (en) | Archive scanning image restoration method, archive scanning image restoration device and storage medium | |
| CN106296688B (en) | Image blur detection method and system based on overall situation estimation | |
| CN113052234A (en) | Jade classification method based on image features and deep learning technology | |
| CN113743413B (en) | Visual SLAM method and system combining image semantic information | |
| CN117726837A (en) | A nonlinear optimization feature matching method | |
| CN117474916A (en) | Image detection method, electronic 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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right | Effective date of registration:20201023 Address after:130000 Changchun Qianjin Street, Jilin, No. 2699 Applicant after:Jilin University Applicant after:Jilin Yugong Intelligent Technology Co.,Ltd. Address before:130012 Chaoyang District Qianjin Street, Jilin, China, No. 2699, No. Applicant before:Jilin University | |
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information | Inventor after:Zhang Ximing Inventor after:Liu Xianggang Inventor after:Liu Weiping Inventor after:Hu Bochun Inventor after:Meng Jin Inventor after:Liu Jia Inventor after:Wang Zhaofeng Inventor after:Wang Yulin Inventor after:Guo Yufeng Inventor after:Guan Zhe Inventor before:Liu Weiping Inventor before:Liu Xianggang Inventor before:Guan Zhe Inventor before:Hu Bochun Inventor before:Meng Jin Inventor before:Liu Jia Inventor before:Wang Zhaofeng Inventor before:Wang Yulin Inventor before:Guo Yufeng Inventor before:Zhang Ximing |