CN111639638B - System and method for identifying transparent flat plate - Google Patents
System and method for identifying transparent flat plate Download PDFInfo
- Publication number
- CN111639638B CN111639638B CN202010479288.0A CN202010479288A CN111639638B CN 111639638 B CN111639638 B CN 111639638B CN 202010479288 A CN202010479288 A CN 202010479288A CN 111639638 B CN111639638 B CN 111639638B
- Authority
- CN
- China
- Prior art keywords
- image
- polarization
- algorithm
- low
- brightness image
- 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
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000010287 polarization Effects 0.000 claims abstract description 66
- 230000001629 suppression Effects 0.000 claims abstract description 30
- 238000005259 measurement Methods 0.000 claims abstract description 29
- 238000004364 calculation method Methods 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 8
- 238000005286 illumination Methods 0.000 claims description 4
- 230000003287 optical effect Effects 0.000 claims description 4
- 238000002493 microarray Methods 0.000 claims description 3
- 125000004432 carbon atom Chemical group C* 0.000 claims description 2
- 150000001875 compounds Chemical class 0.000 claims description 2
- 238000011524 similarity measure Methods 0.000 claims description 2
- 239000011521 glass Substances 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- NIXOWILDQLNWCW-UHFFFAOYSA-N acrylic acid group Chemical group C(C=C)(=O)O NIXOWILDQLNWCW-UHFFFAOYSA-N 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000012780 transparent material Substances 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/145—Illumination specially adapted for pattern recognition, e.g. using gratings
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Studio Devices (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention aims to provide a system for identifying a transparent flat plate, which comprises a polarization camera module, a reflection suppression algorithm module, a similarity measurement algorithm module and an identification module. The invention also discloses a method for identifying the transparent flat plate, which comprises the following steps: taking in light, namely taking in the light of the transparent flat plate by adopting a polarization camera; guiding a polarization image shot by a polarization camera into a computer, and judging the brightness condition of the polarization image through a reflection suppression algorithm; carrying out similarity measurement algorithm on the image output by the reflection suppression algorithm in a computer; and judging whether the transparent flat plate exists or not according to the result of the similarity measurement algorithm. According to the invention, whether the specified area has the transparent flat plate area can be rapidly judged indoors through the reflection suppression algorithm and the similarity measurement algorithm of the computer, and the method is convenient and rapid without excessively complicated hardware equipment.
Description
Technical Field
The invention relates to the field of image recognition technology identification, and provides a system and a method for identifying a transparent flat plate.
Background
With the development of industry, the demand of industrial robots is rapidly increasing, and simultaneously, users have higher requirements on the autonomous navigation capability of the robots. Current visual navigation technology has the ability to assist driving, but lacks an effective identification method for transparent doors or windows in a daily life scenario.
For the identification of transparent objects, a currently common method is to use a data-driven deep learning or a traditional image identification technology, the former needs to manually acquire and process a large amount of data and expensive computer hardware, and the latter needs to manually design an algorithm for a certain kind of scenes, and the designed algorithm has poor adaptability and high labor cost. Compared with a small transparent object with a complex shape and a remarkable shadow and scattering characteristic, the flat transparent flat plate has almost no characteristic, and a bright reflection light spot can be identified only under strong light irradiation, so that the small transparent object is identified by the conventional method. For example, CN109903323A uses a data-driven deep learning method for identifying transparent objects, relies on a large number of pairs of color and depth data and a high-performance processor running algorithms; the patent CN102753933A needs a complex hardware system, the identified transparent body needs to be fixed on the system, and the identification algorithm cannot be used in indoor scenes.
The previous methods are limited by the image sensor and recognition algorithm used and cannot take advantage of the physical properties of the transparent flat object. Therefore, the invention provides a solution for identifying an indoor scene transparent flat plate, which mainly uses a polarization camera to collect a polarization image, uses a computer to inhibit a reflection image in the polarization image and measure the similarity of the image, and judges whether a transparent flat plate exists in a target area.
Disclosure of Invention
The present invention addresses the deficiencies of the prior art and the need for practical production by providing a system for identifying a transparent plate, the system comprising:
a polarization camera module which takes externally-taken light and generates a polarization image;
a reflection suppression algorithm module, receiving the polarization image and generating a low brightness image and a high brightness image program respectively;
the similarity measurement algorithm module compares the low-brightness image program with the high-brightness image program to generate a measurement result;
and the identification module receives the measurement result and compares the measurement result with the target area to obtain identification data.
A method of identifying a transparent plate, the method comprising the steps of:
s1, taking in light, namely taking in the light of a transparent flat plate by adopting a polarization camera;
s2, importing the polarization image shot by the polarization camera into a computer, outputting a low-brightness image and a high-brightness image through a reflection suppression algorithm, and judging the brightness condition of the polarization image;
s3, performing a similarity measurement algorithm on the images output by the reflection suppression algorithm in a computer, and calculating the similarity of the two images by using the similarity measurement algorithm;
and S4, judging whether a transparent flat plate exists or not according to the result of the similarity measurement algorithm.
Further, in step S1, an angle between the optical axis of the polarization camera and the region of the potentially transparent plate is required to be 21 ° to 46 °, so as to avoid the camera from making a 90 ° angle with respect to the target region, and the transparent plate is made of glass or acrylic transparent material.
Furthermore, one side of the camera in the target area of the transparent flat plate needs to have enough illumination conditions, and the lighting is sufficient in the daytime or the lamp source is supplemented at night, so that the camera can capture the light reflected by the transparent flat plate.
Further, the sensor of the polarization camera is selected from a sensor with a microarray polarization filter.
Further, the camera sensor is a sony IMX250MZR.
Further, in step S2, the reflection suppression algorithm is as follows: in the polarization image shot by the polarization camera 2, the values I of every four adjacent pixel points 0 I 45 I 90 I 135 Has the following relations:
I 0 -I(0) ②
I 45 -I(45) ③
I 90 -I(45) ④
I 135 ＝I(135) ⑤
in the above formula, the first and second carbon atoms are,
the angle is expressed, the A, B and C values of (1) can be obtained by using the least square method for (2), (3), (4) and (5), and the low-brightness image and the high-brightness image output by the reflection suppression algorithm are respectively recorded as ^ based on the angle>
Each pixel point in the image is calculated by the following formula:
because of the fact that
Is calculated from every fourth pixel of the polarization image and is therefore one half the size of the polarization image, the pixel being one quarter of the polarization image.
Further, the similarity measure algorithm in step S3 is as follows: inputting a low-brightness image and a high-brightness image output by a reflection suppression algorithm, synchronously sliding windows of the low-brightness image and the high-brightness image along the length direction and the width direction, respectively calculating histograms of the images in the windows of the low-brightness image and the high-brightness image, calculating correlation coefficients of the two histograms as a sliding position result, keeping a position relation of a calculation result of each position of the sliding window, and storing the calculation result in a matrix, wherein the matrix is a measurement result.
Further, the size and step size of the sliding window is selected by the user, and may be one-fourth to sixteen times the size of the image, while the step size may be one-sixteenth to sixty-fourth of the size of the image.
The invention has the following beneficial effects:
the traditional technical scheme adopts a data-driven deep learning method, needs a large amount of paired color and depth data and a high-performance processor for operating an algorithm, is high in implementation difficulty or needs a complex hardware system, an identified transparent body needs to be fixed on the system, and the identification algorithm cannot be used for an indoor scene.
Drawings
FIG. 1 is a flow chart of the overall scheme of the present invention;
FIG. 2 is a Sony IMX250MZR polarization image sensor;
FIG. 3 is a diagram illustrating the correspondence between the polarization image and the high or low brightness image pixels;
fig. 4 is a schematic diagram of a similarity calculation process.
Fig. 5 is a schematic block diagram of a system for identifying a transparent plate.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
With reference to fig. 5, a system for identifying a transparent plate, the system comprising:
a polarization camera module which takes externally-taken light and generates a polarization image;
a reflection suppression algorithm module, which receives the polarization image and generates a low-brightness image and a high-brightness image program respectively;
the similarity measurement algorithm module compares the low-brightness image program with the high-brightness image program to generate a measurement result;
and the identification module receives the measurement result and compares the measurement result with the target area to obtain identification data.
Example 1
Referring to fig. 1, a system and a method for identifying a transparent flat plate are provided, wherein the method comprises the following steps:
s1, taking in light, namely taking in the light of a transparent flat plate by adopting a polarization camera; wherein the optical axis of the polarization camera forms an angle of 21 degrees to 46 degrees with the region of the potential transparent flat plate, the camera is prevented from forming 90 degrees just facing the target region, the transparent flat plate is made of glass or acrylic transparent materials, and the illumination condition is as follows: one side of the camera in the target area of the transparent flat plate needs to have enough illumination conditions, and the lighting is sufficient in the daytime or the lamp source is supplemented at night, so that the camera can capture the light reflected by the transparent flat plate.
S2, guiding the polarization image shot by the polarization camera into a computer, and judging the brightness condition of the polarization image through a reflection suppression algorithm, wherein a sensor with a microarray polarization filter is selected as a sensor of the polarization camera as shown in FIG. 2, such as Sony IMX250MZR, and the reflection suppression algorithm is as follows: referring to fig. 3, in the polarization image captured by the polarization camera, the values I of every four adjacent pixel points C I 45 I 90 I 135 Has the following relations:
I 0 ＝I(0) ②
I 45 ＝I(45) ③
I 90 ＝I(45) ④
I 135 ＝I(135) ⑤
in the above-mentioned formula, the compound has the following structure,
the angle is expressed, the A, B and C values of (1) can be obtained by using the least square method for (2), (3), (4) and (5), and the low-brightness image and the high-brightness image output by the reflection suppression algorithm are respectively recorded as ^ based on the angle>
Each pixel point in the image is calculated by the following formula:
because of the fact that
Is calculated from every fourth pixel of the polarization image and is therefore one half the size of the polarization image, the pixel being one quarter of the polarization image.
S3, performing a similarity measurement algorithm on the image output by the reflection suppression algorithm in a computer, wherein the similarity measurement algorithm is as follows: inputting a low-brightness image and a high-brightness image output by a reflection suppression algorithm, then synchronously sliding windows along the length direction and the width direction for the two images, respectively calculating histograms of the images in the two windows, calculating correlation coefficients of the two histograms as a sliding position result, keeping a position relation of a calculation result of each position of the sliding window, and storing the calculation result in a matrix, wherein the matrix is a measurement result, as shown in fig. 4.
Further, the size and step size of the sliding window is selected by the user, and may be one-fourth to sixteen times the size of the image, and the step size may be one-sixteenth to sixty-fourth of the size of the image.
And S4, judging whether a transparent flat plate exists or not according to the result of the similarity measurement algorithm.
Example 2
The reflection suppression algorithm has various types and can be used for replacing or integrating the method provided by the scheme; if an algorithm using a color image or a gray scale image as input is used, the polarization camera in the scheme can be replaced by a color high-definition camera or a black and white camera, and other methods such as calculating structural similarity SSIM and the like can also be used for the similarity measurement algorithm.
The key point of the invention is that the characteristic that the transparent flat plate can reflect light is utilized, the low-brightness image with the reflected light suppressed and the high-brightness image without the reflected light suppressed are compared, and if the difference between the low-brightness image with the reflected light suppressed and the high-brightness image with the reflected light not suppressed is larger, the existence of the transparent flat plate is judged.
The method comprises the steps of firstly, acquiring a polarization image by using a polarization camera, and easily acquiring a low-brightness image and a high-brightness image by using the physical characteristics of glass without a complex algorithm.
And secondly, the two images are compared, so that the method is visual and easy to explain, and the calculation speed is high. Due to the modular design of the system, the iterative upgrade of the system is simpler. Specifically, the method comprises the following steps: the polarization camera can select different configurations or brands according to requirements; if a more efficient and accurate reflection suppression method exists, the reflection suppression algorithm can be selected; if the reflection suppression algorithm is an intelligent algorithm independent of physical characteristics, the polarization camera can be changed into a common color camera; the similarity measurement algorithm can adopt a similarity measurement method except the histogram correlation coefficient, and can also adopt a plurality of methods to comprehensively measure;
the method can be used not only in an indoor environment, but also in a target area with a distance of 10 meters under the condition of proper angle and light (see the description of the light in the technical description), for example, whether a glass door or window of a nearby building is opened or not can be judged.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (6)
1. A system for identifying a transparent plate, the system comprising:
the polarization camera module is used for taking externally-shot light and generating a polarization image;
the reflection suppression algorithm module is used for receiving the polarization image and outputting a low-brightness image and a high-brightness image through a reflection suppression algorithm;
the similarity measurement algorithm module is used for calculating the similarity of the two images by using a similarity measurement algorithm to obtain a similarity result;
the identification module receives the similarity result and compares the similarity result with the target area to obtain identification data, and judges whether the transparent flat plate exists in the target area or not according to the identification data;
the reflection suppression algorithm is as follows: in the polarization image shot by the polarization camera 2, the values I of every four adjacent pixel points 0 、I 45 、I 90 、I 135 Has the following relations:
I 0 ＝I(0)②
I 45 ＝I(45)③
I 90 ＝I(45)④
I 135 ＝I(135)⑤
in the above formula, the first and second carbon atoms are,
the angle is expressed, the A, B and C values of (1) can be obtained by using the least square method for (2), (3), (4) and (5), and the low-brightness image and the high-brightness image output by the reflection suppression algorithm are respectively represented as I low And I high Each pixel point in the image is calculated by the following formula:
because of I low And I high Is calculated from every fourth pixel of the polarization image, so its width and height dimensions are respectively half of the polarization image, and the pixels are a quarter of the polarization image;
the similarity metric algorithm is as follows: inputting a low-brightness image and a high-brightness image output by a reflection suppression algorithm, synchronously sliding windows of the low-brightness image and the high-brightness image along the length direction and the width direction, respectively calculating histograms of the images in the windows of the low-brightness image and the high-brightness image, calculating correlation coefficients of the two histograms as sliding position results, keeping the position relation of the calculation result of each position of the sliding window, and storing the calculation result in a matrix, wherein the matrix is a measurement result;
the size and step size of the sliding window is selected by the user and is one-fourth to sixteen times the size of the image, while the step size is one-sixteenth to sixty-fourth of the size of the image.
2. A method of identifying a transparent plate, the method comprising the steps of:
s1, taking in light rays, namely taking in the light rays by adopting a polarization camera;
s2, guiding the polarization image shot by the polarization camera into a computer, outputting a low-brightness image and a high-brightness image through a reflection suppression algorithm, and judging the brightness condition of the polarization image;
s3, performing a similarity measurement algorithm on the two images output by the reflection suppression algorithm in a computer, and calculating the similarity of the two images by using the similarity measurement algorithm;
s4, judging whether a transparent flat plate exists in the target area or not according to the result of the similarity measurement algorithm;
the reflection suppression algorithm in step S2 is as follows: in the polarization image shot by the polarization camera 2, the values I of every four adjacent pixel points 0 、I 45 、I 90 、I 135 Has the following relations:
I 0 ＝I(0)②
I 45 ＝I(45)③
I 90 ＝I(45)④
I 135 ＝I(135)⑤
in the above-mentioned formula, the compound has the following structure,
the angle is expressed, the A, B and C values of (1) can be obtained by using the least square method for (2), (3), (4) and (5), and the low-brightness image and the high-brightness image output by the reflection suppression algorithm are respectively represented as I low And I high Each pixel point in the image is calculated by the following formula:
because of I low And I high Is calculated from every fourth pixel of the polarization image, so its width and height dimensions are respectively half of the polarization image, and the pixels are a quarter of the polarization image;
the similarity measure algorithm in step S3 is as follows: inputting a low-brightness image and a high-brightness image output by a reflection suppression algorithm, synchronously sliding windows of the low-brightness image and the high-brightness image along the length direction and the width direction, respectively calculating histograms of the images in the windows of the low-brightness image and the high-brightness image, calculating correlation coefficients of the two histograms as sliding position results, keeping the position relation of the calculation result of each position of the sliding window, and storing the calculation result in a matrix, wherein the matrix is a measurement result;
the size and step size of the sliding window is selected by the user and is one-fourth to sixteen times the size of the image, while the step size is one-sixteenth to sixty-fourth of the size of the image.
3. A method according to claim 2, characterized in that: in step S1, an angle between the optical axis of the polarization camera and the region of the potentially transparent flat plate is required to be 21 ° to 46 °, and a 90 ° angle between the optical axis of the polarization camera and the plane of the target region is avoided.
4. A method according to claim 3, characterized in that: one side of the camera in the target area of the transparent flat plate needs to have enough illumination conditions, and the lighting is sufficient in the daytime or the lamp source is supplemented at night, so that the camera can capture the light reflected by the transparent flat plate.
5. A method according to claim 2, characterized in that: the sensor of the polarization camera adopts a sensor with a microarray polarization filter.
6. The method according to claim 5, characterized in that: the camera sensor is a sony IMX250MZR.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010479288.0A CN111639638B (en) | 2020-05-29 | 2020-05-29 | System and method for identifying transparent flat plate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010479288.0A CN111639638B (en) | 2020-05-29 | 2020-05-29 | System and method for identifying transparent flat plate |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111639638A CN111639638A (en) | 2020-09-08 |
| CN111639638B true CN111639638B (en) | 2023-04-18 |
Family
ID=72332997
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010479288.0A Active CN111639638B (en) | 2020-05-29 | 2020-05-29 | System and method for identifying transparent flat plate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111639638B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113902804B (en) * | 2021-10-14 | 2024-10-25 | 北京达佳互联信息技术有限公司 | Method, device, equipment and medium for identifying transparent area in image |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002098650A (en) * | 2000-09-26 | 2002-04-05 | Matsushita Electric Works Ltd | Transparent body detection method and system thereof |
| EP2889807A2 (en) * | 2013-12-24 | 2015-07-01 | Ricoh Company, Ltd. | Image processing device, image processing system, mobile object control apparatus, image processing method, and carrier means |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060285207A1 (en) * | 2001-01-09 | 2006-12-21 | Hay Ranald J | Techniques for controlling observed glare using polarized optical transmission and reception devices |
| CA2596626C (en) * | 2005-02-02 | 2012-01-24 | Intergraph Software Technologies Company | Real-time image detection using polarization data |
| CN102782720B (en) * | 2009-12-25 | 2015-08-05 | 株式会社理光 | Object recognition equipment, mobile agent control device and information provider unit |
| US10761398B2 (en) * | 2018-06-29 | 2020-09-01 | Mitutoyo Corporation | Imaging ellipsometer system utilizing a tunable acoustic gradient lens |
-
2020
- 2020-05-29 CN CN202010479288.0A patent/CN111639638B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002098650A (en) * | 2000-09-26 | 2002-04-05 | Matsushita Electric Works Ltd | Transparent body detection method and system thereof |
| EP2889807A2 (en) * | 2013-12-24 | 2015-07-01 | Ricoh Company, Ltd. | Image processing device, image processing system, mobile object control apparatus, image processing method, and carrier means |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111639638A (en) | 2020-09-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109376591B (en) | Ship target detection method for deep learning feature and visual feature combined training | |
| CN109871902B (en) | SAR small sample identification method based on super-resolution countermeasure generation cascade network | |
| CN112116633B (en) | Mine drilling counting method | |
| CN116188328B (en) | Parking area response lamp linked system based on thing networking | |
| CN111401246B (en) | Smoke concentration detection method, device, equipment and storage medium | |
| CN1581231A (en) | Infra-red and visible light dynamic image interfusion method based on moving target detection | |
| CN112949452B (en) | Robot low-light environment grabbing detection method based on multitask shared network | |
| CN109389569B (en) | Monitoring video real-time defogging method based on improved DehazeNet | |
| CN107403451B (en) | Self-adaptive binary characteristic monocular vision odometer method, computer and robot | |
| CN113159043A (en) | Feature point matching method and system based on semantic information | |
| CN111639638B (en) | System and method for identifying transparent flat plate | |
| CN111242026A (en) | Remote sensing image target detection method based on spatial hierarchy perception module and metric learning | |
| CN117252868A (en) | Direct current screen defect detection method based on machine vision | |
| CN109766748B (en) | Pedestrian re-recognition method based on projection transformation and dictionary learning | |
| CN114022820A (en) | Intelligent beacon light quality detection method based on machine vision | |
| CN117214172A (en) | Method and device for detecting defects of inner wall of long barrel cylinder and storage medium | |
| CN116385353B (en) | Camera module abnormality detection method | |
| CN113344865A (en) | Method, device, equipment and medium for detecting surface defects of smooth object | |
| CN111898525B (en) | Construction method of smoke identification model, and method and device for detecting smoke | |
| CN118115947A (en) | Cross-mode pedestrian re-identification method based on random color conversion and multi-scale feature fusion | |
| Wu et al. | Identifying color in motion in video sensors | |
| Jia et al. | A novel moving cast shadow detection of vehicles in traffic scene | |
| CN116342644A (en) | Intelligent monitoring method and system suitable for coal yard | |
| CN112633372B (en) | Light source estimation method and device of AR (augmented reality) equipment | |
| CN112926380B (en) | Novel underwater laser target intelligent recognition system |
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 |