CN111026902B - Intelligent identification system and method for building material category - Google Patents
Intelligent identification system and method for building material category Download PDFInfo
- Publication number
- CN111026902B CN111026902B CN201911330456.3A CN201911330456A CN111026902B CN 111026902 B CN111026902 B CN 111026902B CN 201911330456 A CN201911330456 A CN 201911330456A CN 111026902 B CN111026902 B CN 111026902B
- Authority
- CN
- China
- Prior art keywords
- building material
- module
- matching
- identification
- recognition
- 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; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5862—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
- G06Q10/0875—Itemisation or classification of parts, supplies or services, e.g. bill of materials
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
- G06T7/41—Analysis of texture based on statistical description of texture
- G06T7/44—Analysis of texture based on statistical description of texture using image operators, e.g. filters, edge density metrics or local histograms
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Library & Information Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the field of intelligent identification, in particular to an intelligent identification system and method for building material categories, comprising the following steps: the acquisition module is used for acquiring images of building materials; the preprocessing module is used for preprocessing the acquired image; the extraction module is used for extracting texture features of the preprocessed image; the identification module is used for carrying out identification according to the matching of the extracted texture characteristics with the database; and the output module is used for outputting the identification result. According to the invention, by extracting the texture features, the influence of illumination in the process of identifying the building materials is reduced, and the identification effect under the condition of severe illumination change is improved.
Description
Technical Field
The invention relates to the field of intelligent identification, in particular to an intelligent identification system and method for building material categories.
Background
At present, real estate is rapidly developing, and the demand for building materials is also increasing. In order to ensure the supply and sales of building materials, large building material enterprises are equipped with large enough warehouses. Large building material warehouses store a variety of building materials, and it is difficult for the manager to keep up with the production speed of the enterprise by manual inspection and classification, which also brings great working pressure to the staff. In this regard, document CN108229370a discloses a warehouse cargo intelligent recognition system based on cargo image recognition technology, which comprises a data input module, a data storage module, an analysis and comparison module, a phase acquisition module, an output module and an abnormality alarm module. The invention compares the image of the good sample with the image of the actual goods through the image recognition technology to judge whether the detected goods are correct good products, and has high accuracy and high detection speed.
However, image recognition requires collecting pictures of building materials, and the quality of the collected pictures is affected by light and environment, and the accuracy of recognition is directly affected by the quality of the pictures. It can be seen that it is difficult to accurately identify building materials in environments where lighting conditions are poor.
Disclosure of Invention
The invention provides an intelligent recognition system and method for building material types, which solve the technical problem that the existing cargo image recognition system is difficult to accurately recognize building materials in environments with poor illumination conditions.
The basic scheme provided by the invention is as follows: an intelligent identification system for a class of building materials, comprising: the acquisition module is used for acquiring images of building materials; the preprocessing module is used for preprocessing the acquired image; the extraction module is used for extracting texture features of the preprocessed image; the identification module is used for carrying out identification according to the matching of the extracted texture characteristics with the database; and the output module is used for outputting the identification result.
The invention extracts the texture characteristics of the building material when identifying the building material, and carries out graying and binarization processing when extracting the texture; the method reduces the influence of illumination in the process of identifying building materials and improves the identification effect under the condition of severe illumination change.
The intelligent recognition system for building material category recognizes building materials by matching texture features in the image, rather than recognizing by matching the whole image. The influence of the illumination condition on the identification result is greatly eliminated, and the identification accuracy is improved.
Further, the device comprises a storage module for storing the matching and identifying results. Learning samples are provided for matching of texture features so that the system can adaptively analyze new samples of building material to improve accuracy of identification.
Further, the device comprises a learning module for learning according to the history matching and the recognition result. Through learning, optimizing the matching efficiency and accuracy; the error of matching can be corrected, and the accuracy of recognition is improved.
Further, the matching control device comprises a feedback module, and the matching is carried out again when the matching error is larger than a preset threshold value until the matching error is smaller than the preset threshold value. And feeding back the matching result, so that the matching process is continuously corrected, and the recognition accuracy can be improved.
Further, the method comprises a clustering module, wherein the extracted texture features are classified by adopting a k-means clustering algorithm; the method comprises the following specific steps: s11: inputting texture features; s12: randomly selecting K texture features as initial clustering centers; s13: assigning each object to its nearest cluster center; s14: re-calculating a clustering center; s15: if the clustering result is converged, outputting the clustering result; if not, the process returns to step S12. Before matching and recognition, the extracted texture features are classified, so that matching is performed in a certain category, and the probability of matching errors can be reduced.
Further, LBP features are adopted for extracting texture features; the method comprises the following specific steps: s21: obtaining a basic local binary pattern of each pixel of the window; s22: rotation-invariant implementation: circularly right-shifting the binary string to obtain all possible values, and taking the minimum value as the LBP value of the current window; s23: reducing feature dimensions using equivalent patterns; s24: using the LBP value of each pixel to obtain a histogram of the whole window; s25: judging whether to traverse the whole image, and if not: step S21, connecting a step length moving window and returning to the step S21; yes, step S26 is performed; s26: connecting the histograms of all windows to obtain LBP characteristics under the window scale, and scaling the window according to a window scaling factor; s27, judging whether the maximum window upper limit is exceeded, and if so: outputting a result; no: returning to step S21, the image is traversed from the beginning using the new window size. LBP (Local Binary Pattern ) is an operator used to describe the local texture features of an image, has rotation invariance and gray invariance, and has better recognition effect under the condition of intense illumination variation.
Further, a sequential similarity detection algorithm is adopted during matching; the method comprises the following specific steps: s31: selecting an error criterion; s32: setting a constant threshold; s33: randomly selecting one point in the sub-image, calculating the absolute error value of the point and the corresponding point in the template, accumulating the error of each random point pair, stopping accumulation if the error exceeds a set threshold value when the error is accumulated to the r-th time, and recording the accumulation times r at the moment; s34: and calculating an error e for the whole image to obtain a curved surface formed by r values, wherein the position corresponding to the maximum value of the curved surface is the optimal matching position of the template. The sequential similarity detection algorithm looks for a corresponding or similar module in another image (search graph) based on a known image module (template graph). The primary search is performed first, then the fine search is performed, the search range is gradually reduced, and the identification accuracy is greatly improved.
Further, the preprocessing module adopts a Gaussian filtering algorithm. Most of the noise of the image belongs to Gaussian noise, the algorithm is to carry out weighted average on the whole image, and the value of each pixel point is obtained by carrying out weighted average on the pixel point and other pixel values in the neighborhood, so that the method is suitable for eliminating Gaussian noise.
Further, the acquisition module is a CMOS image sensor. The sensor has low power consumption, high imaging speed and wide response range, and is favorable for rapidly collecting images of building materials.
Drawings
FIG. 1 is a block diagram of a system architecture of an embodiment of the intelligent identification system and method for building material categories of the present invention.
Detailed Description
The following is a further detailed description of the embodiments:
example 1
An embodiment of an intelligent identification system for building material categories according to the present invention is generally shown in fig. 1, and includes: the device comprises an acquisition module, a preprocessing module, an extraction module, a clustering module, an identification module, a feedback module, a storage module, a learning module and an output module.
The acquisition module is a CMOS image sensor that acquires texture on the building material and sends the image to the preprocessing module. And after the preprocessing module receives the image, denoising the image by adopting a Gaussian filtering algorithm, and sending the denoised image to the feature extraction module.
And after the feature extraction module receives the image, extracting LBP texture features of the image. The method comprises the following specific steps: step one, a basic local binary pattern of each pixel of the window is obtained. And step two, circularly right-shifting the binary string to obtain all possible values, and taking the minimum value as the LBP value of the current window, thereby realizing rotation invariance. And thirdly, reducing the feature dimension by using an equivalent mode. And step four, utilizing the LBP value of each pixel to obtain a histogram of the whole window. Step five, judging whether to traverse the whole image: if not, connecting a step length moving window, and returning to the step one; if yes, executing the step six. And step six, connecting the histograms of all windows to obtain LBP characteristics under the window scale, and scaling the windows according to window scaling factors. Step seven, judging whether the maximum window upper limit is exceeded or not: if yes, outputting a result; if no: returning to step one, the image is traversed from scratch using the new window size. And after the texture features of the image are extracted, sending the texture features to a clustering module.
And after the clustering module receives the texture features, classifying the extracted texture features by adopting a k-means clustering algorithm. The method comprises the following specific steps: step one, inputting texture features. And step two, randomly selecting K texture features as initial clustering centers. Step three, each object is assigned to the cluster center closest to it. And step four, recalculating the clustering center. Fifthly, if convergence is achieved, outputting a clustering result; if not, returning to the step two. And finally, the clustering module classifies the texture features into a certain class according to the similarity of the texture features, and sends the classification result to the identification module.
The identification module is used for matching the class of the texture features with pre-stored texture features in a database of the class. When matching, adopting a sequential similarity detection algorithm, and searching a corresponding or similar module in another image (search graph) according to a known image module (template graph); the primary search is performed first, then the fine search is performed, and the search range is gradually reduced. The method comprises the following specific steps: step one, an error criterion is selected. If the matching error is not more than 5%, the calculation of the mismatch point is terminated. Step two, a constant threshold value, such as 10%, is set. And thirdly, randomly selecting one point in the sub-image, calculating the absolute error value of the point and the corresponding point in the template, accumulating the error of each random point pair, stopping accumulation if the error exceeds a set threshold value when the error is accumulated to the r-th time, and recording the accumulation times r at the moment. And step four, calculating an error e for the whole image to obtain a curved surface formed by r values, wherein the position corresponding to the maximum value of the curved surface is the optimal matching position of the template. After the matching is completed, the identification result is sent to an output module, and the output module outputs the identification result to a display screen.
Example 2
The only difference from example 1 is that: the system also comprises a storage module, a learning module and a feedback module. The storage module is used for storing the matching and the recognition results and providing a learning sample for matching the texture characteristics so that the system can adaptively analyze the new sample of the building material, thereby improving the recognition accuracy. The learning module extracts the matching and identifying results in the storage module to learn, and optimizes the matching efficiency and accuracy. And through learning, the matching error is corrected, and the recognition accuracy is improved. The feedback module corrects the matching process through negative feedback. And when the error of the matching is larger than a preset threshold value, the matching is carried out again until the error is smaller than the preset threshold value. Thus, the matching result is fed back, so that the matching process is continuously corrected, and the recognition accuracy can be improved.
Example 3
The only difference from example 1 is that: also included are pressure sensors and 3D cameras. The pressure sensor is placed under the building material and connected to the identification module via a wireless network. The pressure sensor is used for measuring the weight of the building material and transmitting the measured weight data of the building material to the identification module through a wireless network. The 3D camera is installed in a warehouse for storing building materials and is connected with the identification module through a wireless network. The 3D camera is used for shooting the length, width and height dimensions of the building material and sending the shot length, width and height dimension data of the building material to the identification module (refer to a high-precision camera calibration algorithm in a 3D measurement system). In addition, a pressure sensor is mounted on the vehicle, which sensor acquires a time-dependent profile of the total weight of the vehicle and the transported goods and transmits the profile to the identification module.
When the identification module receives the length, width and height dimension data and the weight data of the building material, the volume data of the building material is calculated according to the length, width and height dimension data, and then the density value of the building material is calculated according to the weight data and the volume data. Finally, matching is carried out according to the density value, and the type of the building material is identified. Because the density difference of different building materials is very big, the kind of building materials is discerned through matching density value, and the rate of accuracy is very big improvement.
Meanwhile, the recognition module can calculate the change curve of the density value along with time by combining the change curve of the total weight along with time and the volume data. As the warehouse entry and exit positions are basically provided with the deceleration bank, the situation that a driver brakes or starts can also occur. Due to the action of inertial force, the total weight of the transport means and the transported goods obtained by the pressure sensor can fluctuate with time to a certain extent.
Thus, the density value calculated from the formula for amplitude in vibration mechanics fluctuates somewhat over time, but this amplitude is usually small, for example, within 0.2%. When the recognition module detects that the peak of the density value in the oscillation state exceeds a preset threshold value (such as 0.5%), the fluctuation of the density value is not caused by inertia force, and a person can take out or unpack the goods privately in the warehouse.
If the result of the texture feature recognition is different from the result of the density value recognition, or the peak of the density value in the oscillation state is detected to exceed a preset threshold value, the recognition module carries out risk notification to prompt that the building material has a problem. The building material may be counterfeited or may be left behind. In fact, it is common for building materials to be wrapt off and be counterfeited with news.
Therefore, the identification is carried out through the texture features and the density values, and whether the wave peak of the density value in the oscillation state exceeds a preset threshold value is detected, so that the accuracy of building material identification can be improved, and the building material can be effectively prevented from being counterfeited and unpacked.
The foregoing is merely an embodiment of the present invention, and a specific structure and characteristics of common knowledge in the art, which are well known in the scheme, are not described herein, so that a person of ordinary skill in the art knows all the prior art in the application day or before the priority date of the present invention, and can know all the prior art in the field, and have the capability of applying the conventional experimental means before the date, so that a person of ordinary skill in the art can complete and implement the present embodiment in combination with his own capability in the light of the present application, and some typical known structures or known methods should not be an obstacle for a person of ordinary skill in the art to implement the present application. It should be noted that modifications and improvements can be made by those skilled in the art without departing from the structure of the present invention, and these should also be considered as the scope of the present invention, which does not affect the effect of the implementation of the present invention and the utility of the patent. The protection scope of the present application shall be subject to the content of the claims, and the description of the specific embodiments and the like in the specification can be used for explaining the content of the claims.
Claims (10)
1. A intelligent identification system for building material class, its characterized in that: comprising the following steps:
the acquisition module is used for acquiring images of building materials;
the preprocessing module is used for preprocessing the acquired image;
the extraction module is used for extracting texture features of the preprocessed image;
the identification module is used for carrying out identification according to the matching of the extracted texture characteristics with the database;
the output module is used for outputting the identification result;
the system also comprises a pressure sensor and a 3D camera, wherein the pressure sensor is arranged under the building material and is connected with the identification module through a wireless network, the pressure sensor is used for measuring the weight of the building material and transmitting the measured weight data of the building material to the identification module, the 3D camera is arranged in a warehouse for storing the building material, the 3D camera is connected with the identification module, the 3D camera is used for shooting the length, width and height dimensions of the building material, sending the shot length, width and height dimension data of the building material to the identification module, the pressure sensor is further arranged on the transport means, and the pressure sensor acquires a change curve of the total weight of the transport means and the transported goods along with time and sends the change curve to the identification module;
when the identification module receives the length, width and height dimension data and the weight data of the building material, firstly calculating the volume data of the building material according to the length, width and height dimension data, then calculating the density value of the building material according to the weight data and the volume data, and finally matching according to the density value to identify the type of the building material;
meanwhile, the recognition module can calculate the change curve of the density value along with time by combining the change curve of the total weight along with time and the volume data, and when the recognition module detects that the peak of the density value in the oscillation state exceeds a preset threshold value, the fluctuation of the density value is judged not to be caused by inertia force, but to be a person in the warehouse to unload the goods or to drop the goods;
if the result of the texture feature recognition is different from the result of the density value recognition, or the peak of the density value in the oscillation state is detected to exceed a preset threshold value, the recognition module carries out risk notification to prompt that the type of the building material has a problem.
2. The intelligent recognition system for building material categories of claim 1, wherein: the device comprises a storage module for storing the matching and identifying results.
3. The intelligent recognition system for building material categories of claim 2, wherein: the learning module is used for learning according to the history matching and the recognition result.
4. The intelligent recognition system for building material categories of claim 3, wherein: the matching method comprises a feedback module, and the matching is carried out again when the matching error is larger than a preset threshold value until the matching error is smaller than the preset threshold value.
5. The intelligent recognition system for building material categories of claim 4, wherein: the method comprises a clustering module, wherein the extracted texture features are classified by adopting a k-means clustering algorithm; the method comprises the following specific steps: s11: inputting texture features; s12: randomly selecting K texture features as initial clustering centers; s13: assigning each object to its nearest cluster center; s14: re-calculating a clustering center; s15: if the clustering result is converged, outputting the clustering result; if not, the process returns to step S12.
6. The intelligent recognition system for building material categories of claim 5, wherein: extracting texture features by using LBP features; the method comprises the following specific steps: s21: obtaining a basic local binary pattern of each pixel of the window; s22: rotation-invariant implementation: circularly right-shifting the binary string to obtain all possible values, and taking the minimum value as the LBP value of the current window; s23: reducing feature dimensions using equivalent patterns; s24: using the LBP value of each pixel to obtain a histogram of the whole window; s25: judging whether to traverse the whole image, and if not: step S21, connecting a step length moving window and returning to the step S21; yes, step S26 is performed; s26: connecting histograms of all windows to obtain LBP characteristics under the window scale, and scaling the windows according to window scaling factors; s27: judging whether the maximum window upper limit is exceeded, and if yes: outputting a result; no: returning to step S21, the image is traversed from the beginning using the new window size.
7. The intelligent recognition system for building material categories of claim 6, wherein: a sequential similarity detection algorithm is adopted during matching; the method comprises the following specific steps: s31: selecting an error criterion; s32: setting a constant threshold; s33: randomly selecting one point in the sub-image, calculating the absolute error value of the point and the corresponding point in the template, accumulating the error of each random point pair, stopping accumulation if the error exceeds a set threshold value when the error is accumulated to the r-th time, and recording the accumulation times r at the moment; s34: and calculating an error e for the whole image to obtain a curved surface formed by r values, wherein the position corresponding to the maximum value of the curved surface is the optimal matching position of the template.
8. The intelligent recognition system for building material categories of claim 7, wherein: the preprocessing module adopts a Gaussian filtering algorithm.
9. The intelligent recognition system for building material categories of claim 8, wherein: the acquisition module is a CMOS image sensor.
10. The intelligent identification method for the building material category is characterized by comprising the following steps of: the method comprises the following steps:
s01: collecting an image of the building material;
s02: preprocessing the acquired image, including graying and binarization;
s03: extracting texture features;
s04: clustering texture features;
s05: matching the extracted texture features with a database for identification;
s06: judging whether the matched error is larger than a preset threshold value, and re-matching when the matched error is larger than the preset threshold value until the error is smaller than the preset threshold value;
s07: storing the matching and identifying results;
s08: learning according to the stored matching and recognition results;
s09: outputting a recognition result;
the pressure sensor is arranged below the building material and connected with the identification module through a wireless network, the pressure sensor is used for measuring the weight of the building material and transmitting measured weight data of the building material to the identification module, the 3D camera is arranged in a warehouse for storing the building material, the 3D camera is connected with the identification module, the 3D camera is used for shooting the length, width and height dimensions of the building material and transmitting shot length, width and height dimension data of the building material to the identification module, the pressure sensor is further arranged on a transport means, and the pressure sensor acquires a change curve of the total weight of the transport means and transported goods along with time and transmits the change curve to the identification module;
when the identification module receives the length, width and height dimension data and the weight data of the building material, firstly calculating the volume data of the building material according to the length, width and height dimension data, then calculating the density value of the building material according to the weight data and the volume data, and finally matching according to the density value to identify the type of the building material; meanwhile, the recognition module can calculate the change curve of the density value along with time by combining the change curve of the total weight along with time and the volume data, and when the recognition module detects that the peak of the density value in the oscillation state exceeds a preset threshold value, the fluctuation of the density value is judged not to be caused by inertia force, but to be a person in the warehouse to unload the goods or to drop the goods;
if the result of the texture feature recognition is different from the result of the density value recognition, or the peak of the density value in the oscillation state is detected to exceed a preset threshold value, the recognition module carries out risk notification to prompt that the type of the building material has a problem.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911330456.3A CN111026902B (en) | 2019-12-20 | 2019-12-20 | Intelligent identification system and method for building material category |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911330456.3A CN111026902B (en) | 2019-12-20 | 2019-12-20 | Intelligent identification system and method for building material category |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111026902A CN111026902A (en) | 2020-04-17 |
| CN111026902B true CN111026902B (en) | 2023-06-16 |
Family
ID=70212430
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911330456.3A Active CN111026902B (en) | 2019-12-20 | 2019-12-20 | Intelligent identification system and method for building material category |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111026902B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107748877A (en) * | 2017-11-10 | 2018-03-02 | 杭州晟元数据安全技术股份有限公司 | A kind of Fingerprint recognition method based on minutiae point and textural characteristics |
| CN108805018A (en) * | 2018-04-27 | 2018-11-13 | 淘然视界(杭州)科技有限公司 | Road signs detection recognition method, electronic equipment, storage medium and system |
| CN110119753A (en) * | 2019-01-08 | 2019-08-13 | 长江岩土工程总公司（武汉） | A kind of method of reconstituted texture identification lithology |
| CN110136160A (en) * | 2019-05-13 | 2019-08-16 | 南京大学 | A kind of rapid image matching method based on circular projection |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102542303A (en) * | 2010-12-24 | 2012-07-04 | 富士通株式会社 | Device and method for generating classifier of specified object in detection image |
| US9418446B2 (en) * | 2014-08-27 | 2016-08-16 | Nokia Technologies Oy | Method and apparatus for determining a building location based on a building image |
| US9588098B2 (en) * | 2015-03-18 | 2017-03-07 | Centre De Recherche Industrielle Du Quebec | Optical method and apparatus for identifying wood species of a raw wooden log |
-
2019
- 2019-12-20 CN CN201911330456.3A patent/CN111026902B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107748877A (en) * | 2017-11-10 | 2018-03-02 | 杭州晟元数据安全技术股份有限公司 | A kind of Fingerprint recognition method based on minutiae point and textural characteristics |
| CN108805018A (en) * | 2018-04-27 | 2018-11-13 | 淘然视界(杭州)科技有限公司 | Road signs detection recognition method, electronic equipment, storage medium and system |
| CN110119753A (en) * | 2019-01-08 | 2019-08-13 | 长江岩土工程总公司（武汉） | A kind of method of reconstituted texture identification lithology |
| CN110136160A (en) * | 2019-05-13 | 2019-08-16 | 南京大学 | A kind of rapid image matching method based on circular projection |
Non-Patent Citations (2)
| Title |
|---|
| 冉宝山.《基于机器视觉的装料系统试验研究》.《信息科技》.2016,正文第1.2.3-2.3.4节，第3-4章. * |
| 蔺璇.《LBP算法在织物疵点检测及分类中的应用》.《工程科技Ⅰ辑》.2016,正文第2.1，4.4.2节. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111026902A (en) | 2020-04-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3340106B1 (en) | Method and system for assigning particular classes of interest within measurement data | |
| CN110245663B (en) | Method for identifying steel coil information | |
| De Almeida et al. | PKLot–A robust dataset for parking lot classification | |
| US9830704B1 (en) | Predicting performance metrics for algorithms | |
| US10062008B2 (en) | Image based object classification | |
| CN108090406B (en) | Face recognition method and system | |
| CN111461107A (en) | Material handling method, apparatus and system for identifying regions of interest | |
| CN102007499A (en) | Detecting facial expressions in digital images | |
| CN112200121B (en) | Hyperspectral unknown target detection method based on EVM and deep learning | |
| US20130163858A1 (en) | Component recognizing apparatus and component recognizing method | |
| US11676182B2 (en) | Computer vision systems and methods for automatically detecting, classifying, and pricing objects captured in images or videos | |
| US20200005075A1 (en) | Computer Vision Systems and Methods for Automatically Detecting, Classifying, and Pricing Objects Captured in Images or Videos | |
| EP3311333B1 (en) | Pairing of images of postal articles with descriptors of singularities of the gradient field | |
| Moirogiorgou et al. | Intelligent robotic system for urban waste recycling | |
| CN111026902B (en) | Intelligent identification system and method for building material category | |
| CN113947570A (en) | Crack identification method based on machine learning algorithm and computer vision | |
| CN106169086B (en) | High-resolution optical image under navigation data auxiliary damages method for extracting roads | |
| CN117274967A (en) | Multi-mode fusion license plate recognition algorithm based on convolutional neural network | |
| CN111950357A (en) | Marine water surface garbage rapid identification method based on multi-feature YOLOV3 | |
| CN114662594B (en) | Target feature recognition analysis system | |
| CN115082832A (en) | Information identification method, device and storage medium | |
| KR20230150625A (en) | Baggage sorting system | |
| CN114120175A (en) | Method for identifying foreign matters on coal conveying belt based on computer vision | |
| CN113705672A (en) | Threshold value selection method, system and device for image target detection and storage medium | |
| Mohottala et al. | Vehicle class recognition using 3d cg models |
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 | ||
| CP03 | Change of name, title or address |
Address after: Room 2502, Building 3, Xingyi, Jushan Street, Xingyi City, Qianxinan Buyi and Miao Autonomous Prefecture, Guizhou Province, 562400 Patentee after: Guizhou Fantian Technology Co.,Ltd. Address before: 562400 room 602, unit 2, building 25, feiyanghuafu, Fengdu sub district office, Xingyi City, Qianxinan Buyei and Miao Autonomous Prefecture, Guizhou Province Patentee before: Guizhou Qianan Technology Co.,Ltd. |
|
| CP03 | Change of name, title or address |