CN109785319B - System, device and method for detecting code-spraying character based on image processing - Google Patents
System, device and method for detecting code-spraying character based on image processing Download PDFInfo
- Publication number
- CN109785319B CN109785319B CN201910077214.1A CN201910077214A CN109785319B CN 109785319 B CN109785319 B CN 109785319B CN 201910077214 A CN201910077214 A CN 201910077214A CN 109785319 B CN109785319 B CN 109785319B
- Authority
- CN
- China
- Prior art keywords
- character
- image
- workpiece
- image processing
- detected
- 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
Classifications
-
- 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
Abstract
The invention belongs to the technical field of industrial detection, and discloses a code-spraying character detection system, a device and a detection method based on image processing, wherein an industrial camera and a sphere integral light source are fixed; adjusting the focal length of an industrial lens; according to the adjustment of the industrial lens, obtaining the point corresponding to the most clear zoom ring of the picture, and continuously repeating the process around the corresponding point until the picture is adjusted to be the most clear; adjusting the conveying speed of the conveying belt to determine the fixed position of the photographing triggering device; the stepping conveying unit conveys the workpiece to be detected, acquires an image of the workpiece to be detected, and processes the image to obtain a result whether the code spraying character is qualified or not; and rejecting unqualified workpieces through an executing mechanism. The invention combines image processing with industrial detection, adopts a mode of combining image processing with automation to detect whether the code-spraying character is qualified, has simple operation and strong universality, has better real-time performance and accuracy, and can greatly improve the detection efficiency.
Description
Technical Field
The invention belongs to the technical field of industrial detection, and particularly relates to a code-spraying character detection system, a device and a detection method based on image processing.
Background
Currently, the current state of the art commonly used in the industry is as follows:
the code spraying workpiece is very commonly used in the fields of industrial manufacture, assembly and the like. In workpiece manufacturing, the factory workpieces are required to be subjected to character code spraying according to strict standards so as to facilitate subsequent assembly. The traditional code-spraying character detection mode adopts manual visual inspection, which is still the mode adopted by many workpiece factories so far, and the work is monotonous and labor-intensive. Because the traditional detection adopts a manual contact mode, the surface pollution of the workpiece is extremely easy to cause, and the detection result is inconvenient for carrying out the subsequent statistical analysis on the quality of the workpiece product.
In summary, the problems of the prior art are:
(1) In the prior art, in the code-spraying character detection, a method combining image processing and automation is not adopted, so that the production efficiency is low.
(2) The existing character recognition system has poor expandability.
(3) The non-contact mode is not adopted in the measurement, and the workpiece to be detected is easy to pollute. The detection result is inconvenient for subsequent statistical analysis of the quality of the workpiece product.
Meaning of solving the technical problems:
the invention combines image processing with industrial detection, adopts a mode of combining image processing with automation to detect whether the code-spraying character is qualified, outputs the unqualified result to an actuating mechanism, eliminates unqualified workpieces, and completes the code-spraying character detection of the workpieces and the processing of detection results. The invention solves the problems that the existing code-spraying detection product can only output results, but can not carry out comprehensive statistical analysis on scattered results.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a code-spraying character detection system and a method based on image processing.
The invention is realized in such a way that the code-spraying character detection method based on image processing comprises the following steps:
dividing a workpiece area to be detected by a dynamic threshold method, wherein f represents a collected workpiece picture, and f (i, j) represents the pixel value of an ith row and a jth column;
secondly, designing a filtering template, wherein the size of the template is mask width multiplied by mask height, the mask width and the mask height are both positive odd numbers, and the template convolves the image f and divides the template size to obtain a new image f mean The method comprises the steps of carrying out a first treatment on the surface of the Selecting a compensation value T offset For satisfying the formula f (i, j) > f mean (i,j)+T offset The pixel points of (2) are reserved in f, and are discarded otherwise; obtaining an image f dyn F is removed by morphological filtering dyn The shot noise existing in the character region picture f is extracted cr The method comprises the steps of carrying out a first treatment on the surface of the Obtaining character areas with communicated areas;
thirdly, obtaining the center point coordinate C of the character area Original-center (i, j) and an angle alpha with respect to the horizontal direction Original Creating affine transformation matrix M HomMat2DRotate ，f cr And M is as follows HomMat2DRotate Performing convolution operation: f (f) affine ＝f gap *M HomMat2DRotate Obtaining an affine transformed image f affine The character area is vertical to the horizontal direction of the image;
fourth, calculating the connected domain f of the region ori_region By designing the area characteristics S of the region feature Width and height characteristics WH feature Set S of (2) feature ∪WH feature Filtering the stray region to obtain a connected region f las_region ；
Fifthly, loading a pre-trained character classifier to detect the region to be recognized to obtain the confidence delta of the recognized character corresponding to each character i (i∈N + )；
Sixth, according to the confidence delta of the character i Calculating Euclidean distance gamma between the character and the corresponding character in the standard character library i ，i∈N + The method comprises the steps of carrying out a first treatment on the surface of the A threshold value epsilon is set for each character, when adelta i +bγ i ≤ε，i∈N + The character with the subscript of i is qualified, and otherwise, the character with the subscript of i is unqualified; threshold value discrimination is carried out on each character, and when the confidence coefficient and Euclidean of all the charactersAnd if the weighted sum of the distances is not greater than the threshold value, the workpiece is qualified, and otherwise, the workpiece is disqualified.
Further, in the first step, an image enhancement template is created to primarily enhance the image, and the contrast between the region to be identified and the background is increased.
Further, the method for detecting the code-spraying character based on the image processing specifically comprises the following steps:
step one, fixing an industrial camera and a ball integral light source;
step two, adjusting the focal length of the industrial lens to enable the picture to be blurred from clear to blurred;
step three, adjusting the industrial lens according to the step two to obtain a point corresponding to a most clear zooming ring of the picture, and continuously repeating the process near the corresponding point until the picture is adjusted to be the most clear;
and step four, adjusting the conveying speed of the conveying belt according to the result of the step three so as to determine the fixed position of the photographing triggering device.
Step five, the stepping conveying unit conveys the workpiece to be detected, acquires an image of the workpiece to be detected, and processes the image to obtain a result whether the code spraying character is qualified or not;
and step six, rejecting unqualified workpieces through an executing mechanism.
Another object of the present invention is to provide an image processing-based code-spraying character detection system, which is an image acquisition and processing unit; the method specifically comprises the following steps: an industrial camera, an industrial lens, a ball integral light source, an image processing unit and a photographing triggering device;
the photographing triggering device is fixed on the frame; the sphere integral light source is positioned right above the frame conveyor belt; the industrial camera and the industrial lens are positioned right above the ball integral light source; the image processing unit is connected with the industrial camera and is used for processing the image to be detected and outputting a detection result.
Further, the image processing unit obtains a result of whether the workpiece spray code character is qualified or not, and sends a processing instruction to the execution mechanism to reject unqualified workpieces.
Another object of the present invention is to provide an apparatus for detecting a code-spraying character based on image processing, which includes a frame, a step-by-step transmission unit, a photographing trigger device, an image acquisition and processing unit, and an executing mechanism;
the rack is used for placing a workpiece to be detected and supporting other hardware equipment;
the step conveying unit is used for conveying the workpiece to be detected; the photographing triggering device is used for triggering the camera to acquire pictures of the workpiece to be detected;
the image acquisition and processing unit is used for acquiring pictures of the workpiece to be detected and obtaining a processing result; and the actuating mechanism eliminates unqualified workpieces according to the obtained result.
Further, the longitudinal axis of the frame is orthogonal to the central axes of the image acquisition unit and the photographing triggering device of the image acquisition and processing unit.
Further, the step conveying unit comprises a step conveying control unit, a conveying belt and a step motor; the conveying belt is used for conveying the workpiece to be detected, the conveying control unit controls the conveying speed, and the stepping motor is used for providing conveying power.
Further, the photographing triggering device comprises an infrared photoelectric correlation sensing probe and a serial port output device. After triggering the infrared probe, the workpiece to be detected outputs a shooting triggering instruction to an image acquisition unit of the image acquisition and processing unit through a serial port; and an image acquisition unit of the image acquisition and processing unit acquires a picture of the workpiece to be detected.
Another object of the present invention is to provide a code-spraying work piece detecting production line in which the code-spraying character detecting device based on image processing is installed.
In summary, the invention has the advantages and positive effects that:
the detection of human eyes to the code spraying character is 1.2 characters/second, and the invention adopts a method combining image processing and automation in the code spraying character detection to reach 20 characters/second, thereby improving the production efficiency and reducing the labor intensity.
The non-contact mode is adopted for measurement, so that the workpiece to be detected is not polluted.
Simple operation, strong universality and low cost.
The detection result is convenient for carrying out subsequent statistical analysis on the quality of the workpiece product.
Drawings
Fig. 1 is a schematic structural diagram of an apparatus for detecting an inkjet character based on image processing according to an embodiment of the present invention.
In the figure: 1. a frame; 2. a step-by-step conveying unit; 3. a shooting trigger device; 4. an industrial camera; 5. an industrial lens; 6. a sphere integral light source; 7. an image processing unit (industrial personal computer); 8. an actuator.
Fig. 2 is a flowchart of an installation method of an inkjet character detection system based on image processing according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the following examples in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In the prior art, in the detection of the code-spraying character, a method combining image processing and automation is not adopted, so that the production efficiency is low; the non-contact mode is not adopted in the measurement, and the workpiece to be detected is easy to pollute. The detection result is inconvenient for subsequent statistical analysis of the quality of the workpiece product.
In order to solve the above technical problems, the present invention will be described in detail with reference to specific embodiments.
As shown in fig. 1, an apparatus for detecting an inkjet character based on image processing according to an embodiment of the present invention includes:
the device comprises a frame 1, a stepping conveying unit 2, a photographing triggering device 3, an image acquisition and processing unit and an executing mechanism 8; the rack 1 is used for placing a workpiece to be detected and supporting other hardware equipment; the step conveying unit 2 is used for conveying a workpiece to be detected; the photographing triggering device 3 is used for triggering a camera to acquire pictures of the workpiece to be detected; the image acquisition and processing unit is used for acquiring pictures of the workpiece to be detected and obtaining a processing result; and the actuating mechanism 8 eliminates unqualified workpieces according to the obtained result.
The longitudinal axis of the frame 1 is orthogonal to the central axes of the image acquisition unit and the photographing triggering device 3 of the image acquisition and processing unit. The step conveying unit 2 comprises a step conveying control unit, a conveying belt and a step motor; the conveying belt is used for conveying the workpiece to be detected, the conveying control unit controls the conveying speed, and the stepping motor is used for providing conveying power.
The shooting trigger device 3 comprises an infrared photoelectric correlation sensing probe and a serial port output device. After triggering the infrared probe, the workpiece to be detected outputs a shooting triggering instruction to an image acquisition unit of the image acquisition and processing unit through a serial port; and an image acquisition unit of the image acquisition and processing unit acquires a picture of the workpiece to be detected.
The image acquisition and processing unit comprises an industrial camera 4, an industrial lens 5, a sphere integral light source 6, an image processing unit (industrial personal computer) 7 and a photographing triggering device 3; the photographing triggering device 3 is fixed on the frame 1; the sphere integral light source 6 is positioned right above the conveyor belt of the frame 1; the industrial camera 4 and the industrial lens 5 are positioned right above the sphere integral light source 6; the image processing unit 7 is connected with the industrial camera 4, and is used for processing the image to be detected and outputting a detection result.
And the execution mechanism 8 eliminates unqualified workpieces according to the output result of the image processing unit.
The character area to be detected is obtained by an image acquisition unit of an image acquisition and processing unit, and an industrial camera 4 and a sphere integral light source 6 are fixed in the acquisition process; adjusting the focal length of the industrial lens 5 to enable the picture to be blurred to clear and then blurred; adjusting the industrial lens to obtain a point corresponding to the clearest zoom ring of the picture, and continuously repeating the process near the corresponding point until the clearest picture is adjusted; adjusting the conveying speed of the conveying belt to determine the fixed position of the photographing triggering device; the stepping conveying unit conveys the workpiece to be detected and acquires an image of the workpiece to be detected.
Based on the above-mentioned device for detecting the code-spraying character based on image processing, the method for detecting the code-spraying character based on image processing provided by the embodiment of the invention comprises the following steps:
s101: industrial cameras, sphere integral light sources are fixed.
S102: and adjusting the focal length of the industrial lens to enable the picture to be blurred to clear and then blurred.
S103: and step two, adjusting the industrial lens to obtain a point corresponding to the most clear zoom ring of the picture, and continuously repeating the process nearby the corresponding point until the picture is adjusted to be the most clear.
S104: and (3) adjusting the conveying speed of the conveying belt according to the result of the step (III) so as to determine the fixed position of the photographing triggering device.
S105: and the stepping conveying unit conveys the workpiece to be detected, acquires the image of the workpiece to be detected, and processes the image to obtain a result whether the code spraying character is qualified or not.
S106: and rejecting unqualified workpieces through an executing mechanism.
In the embodiment of the present invention, the method for processing the image in step S105 includes:
the first step: dividing a workpiece area to be detected by a dynamic threshold method, wherein f represents an acquired workpiece picture, and f (i, j) represents the pixel value of an ith row and a jth column; firstly, an image enhancement template is created, the image is primarily enhanced, and the contrast between the region to be identified and the background is increased.
And a second step of: designing a filtering template, wherein the size of the template is mask width multiplied by mask height, the mask width and mask height are both positive odd numbers, convoluting the image f by the template, and dividing the image f by the size of the template to obtain a new image f mean The method comprises the steps of carrying out a first treatment on the surface of the Selecting a compensation value T offset For satisfying the formula f (i, j) > f mean (i,j)+T offset The pixel points of (2) are reserved in f, and are discarded otherwise; obtaining an image f dyn ，f dyn The influence of shot noise also exists in the character region image, and the noise is removed through morphological filtering to extract the character region image f cr The method comprises the steps of carrying out a first treatment on the surface of the And obtaining character areas with communicated areas.
And a third step of: obtaining the center point coordinate C of the character area Original-center (i, j) and an angle alpha with respect to the horizontal direction Original Creating affine transformation matrix M based on the above HomMat2DRotate ，f cr And M is as follows HomMat2DRotate Performing convolution operation: f (f) affine ＝f gap *M HomMat2DRotate Obtaining an affine transformed image f affine The character area at this time is perpendicular to the image horizontal direction.
Fourth step: connected domain f of calculation region ori_region At this point there may be stray areas that cannot be filtered out in the second step. Area features S through design area feature Width and height characteristics WH feature Set S of (2) feature ∪WH feature Filtering the stray region to obtain a connected region f las_region 。
Fifth step: loading a pre-trained character classifier to detect a region to be recognized to obtain the confidence delta of the recognized character corresponding to each character i (i∈N + )。
Sixth step: according to confidence level delta of character i Calculating Euclidean distance gamma between the character and the corresponding character in the standard character library i (i∈N + ) The method comprises the steps of carrying out a first treatment on the surface of the A threshold value epsilon is set for each character, when adelta i +bγ i ≤ε(i∈N + ) And if the character with the index of i is qualified, otherwise, the character with the index of i is not qualified. And judging the threshold value of each character, and only if the weighted sum of the confidence coefficient and Euclidean distance of all the characters is not larger than the threshold value, judging the workpiece to be qualified, otherwise, judging the workpiece to be unqualified.
The embodiment of the invention has better recognition effect on the code spraying characters of different workpieces, the detection of the code spraying characters by human eyes is 1.2 characters/second, and the embodiment of the invention adopts a method combining image processing and automation to reach 20 characters/second, thereby improving the production efficiency and reducing the labor intensity.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.
Claims (10)
1. The method for detecting the code spraying character based on the image processing is characterized by comprising the following steps of:
dividing a workpiece area to be detected by a dynamic threshold method, wherein f represents a collected workpiece picture, and f (i, j) represents the pixel value of an ith row and a jth column;
secondly, designing a filtering template, wherein the size of the filtering template is mask width multiplied by mask height, the mask width and the mask height are both positive odd numbers, and the template convolves the image f and divides the size of the template to obtain a new image f mean The method comprises the steps of carrying out a first treatment on the surface of the Selecting a compensation value T offset For satisfying the formula f (i, j) > f mean (i,j)+T offset The pixel points of (2) are reserved in f, and are discarded otherwise; obtaining an image f dyn F is removed by morphological filtering dyn The shot noise existing in the character region picture f is extracted cr The method comprises the steps of carrying out a first treatment on the surface of the Obtaining character areas with communicated areas;
thirdly, obtaining the center point coordinate C of the character area Original-center (i, j) and an angle alpha with respect to the horizontal direction Original Creating affine transformation matrix M HomMat2DRotate ，f cr And M is as follows HomMat2DRotate Performing convolution operation: f (f) affine ＝f gap *M HomMat2DRotate Obtaining an affine transformed image f affine The character area is vertical to the horizontal direction of the image;
fourth, calculating the connected domain f of the region ori_region By designing the area characteristics S of the region feature Width and height characteristics WH feature Set S of (2) feature ∪WH feature Filtering the stray region to obtain a connected region f las_region ；
Fifthly, loading a pre-trained character classifier to detect the region to be recognized to obtain the confidence delta of the recognized character corresponding to each character i (i∈N + )；
Sixth, according to the confidence delta of the character i Calculating Euclidean distance gamma between the character and the corresponding character in the standard character library i ，i∈N + The method comprises the steps of carrying out a first treatment on the surface of the A threshold value epsilon is set for each character, when adelta i +bγ i ≤ε，i∈N + Character with index i is qualified, otherwiseThen the product is disqualified; and judging the threshold value of each character, and if the weighted sum of the confidence coefficient and Euclidean distance of all the characters is not larger than the threshold value, the workpiece is qualified, otherwise, the workpiece is unqualified.
2. The method for detecting an inkjet character based on image processing according to claim 1, wherein in the first step, an image enhancement template is created to perform preliminary enhancement on the image, and the contrast between the region to be identified and the background is increased.
3. The method for detecting an inkjet character based on image processing according to claim 1, wherein the method for detecting an inkjet character based on image processing specifically comprises:
step one, fixing an industrial camera and a ball integral light source;
step two, adjusting the focal length of the industrial lens;
step three, adjusting the industrial lens according to the step two to obtain a point corresponding to a most clear zooming ring of the picture, and continuously repeating the process near the corresponding point until the picture is adjusted to be the most clear;
step four, adjusting the conveying speed of the conveying belt according to the result of the step three to determine the fixed position of the photographing triggering device;
step five, the stepping conveying unit conveys the workpiece to be detected, acquires an image of the workpiece to be detected, and processes the image to obtain a result whether the code spraying character is qualified or not;
and step six, rejecting unqualified workpieces through an executing mechanism.
4. An image processing-based code-spraying character detection system according to the image processing-based code-spraying character detection method of claim 1, wherein the image processing-based code-spraying character detection system is an image acquisition and processing unit; the method specifically comprises the following steps: an industrial camera, an industrial lens, a ball integral light source, an image processing unit and a photographing triggering device;
the photographing triggering device is fixed on the frame; the sphere integral light source is positioned right above the frame conveyor belt; the industrial camera and the industrial lens are positioned right above the ball integral light source; the image processing unit is connected with the industrial camera and is used for processing the image to be detected and outputting a detection result.
5. The image processing-based code spraying character detecting system according to claim 4, wherein the image processing unit obtains a result of whether the code spraying character of the workpiece is qualified or not, and sends a processing instruction to the executing mechanism to reject the unqualified workpiece.
6. An image processing-based code-spraying character detection device using the image processing-based code-spraying character detection method according to claim 1, wherein the image processing-based code-spraying character detection device comprises a frame, a stepping transmission unit, a photographing triggering device, an image acquisition and processing unit and an executing mechanism;
the rack is used for placing a workpiece to be detected and supporting other hardware equipment;
the step conveying unit is used for conveying the workpiece to be detected; the photographing triggering device is used for triggering the camera to acquire pictures of the workpiece to be detected;
the image acquisition and processing unit is used for acquiring pictures of the workpiece to be detected and obtaining a processing result; and the actuating mechanism eliminates unqualified workpieces according to the obtained result.
7. The image processing-based code-spraying character detection device according to claim 6, wherein the longitudinal axis of the frame is orthogonal to the central axes of the image acquisition unit and the photographing trigger device of the image acquisition and processing unit.
8. The image processing-based code-spraying character detection device according to claim 6, wherein the step-by-step conveying unit includes a step-by-step conveying control unit, a conveyor belt, a step motor; the conveying belt is used for conveying the workpiece to be detected, the conveying control unit controls the conveying speed, and the stepping motor is used for providing conveying power.
9. The device for detecting the code spraying character based on the image processing as in claim 6, wherein the shooting trigger device comprises an infrared photoelectric correlation sensing probe and a serial port output device; after triggering the infrared probe, the workpiece to be detected outputs a shooting triggering instruction to an image acquisition unit of the image acquisition and processing unit through a serial port; and an image acquisition unit of the image acquisition and processing unit acquires a picture of the workpiece to be detected.
10. A code-spraying work detection line equipped with the image processing-based code-spraying character detection device of claim 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910077214.1A CN109785319B (en) | 2019-01-28 | 2019-01-28 | System, device and method for detecting code-spraying character based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910077214.1A CN109785319B (en) | 2019-01-28 | 2019-01-28 | System, device and method for detecting code-spraying character based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109785319A CN109785319A (en) | 2019-05-21 |
| CN109785319B true CN109785319B (en) | 2023-09-08 |
Family
ID=66502637
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910077214.1A Active CN109785319B (en) | 2019-01-28 | 2019-01-28 | System, device and method for detecting code-spraying character based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109785319B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110816096A (en) * | 2019-11-18 | 2020-02-21 | 珠海格力智能装备有限公司 | Jet printing processing method and jet printing processing equipment |
| CN111539582B (en) * | 2020-05-11 | 2023-07-21 | 湖北汽车工业学院 | Image processing-based steel plate cutting planning device and method |
| CN113329613B (en) * | 2021-06-16 | 2023-03-24 | 武汉纺织大学 | Chip mounter suction nozzle detection system and method |
| CN113436180A (en) * | 2021-07-07 | 2021-09-24 | 京东科技控股股份有限公司 | Method, device, system, equipment and medium for detecting spray codes on production line |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6553131B1 (en) * | 1999-09-15 | 2003-04-22 | Siemens Corporate Research, Inc. | License plate recognition with an intelligent camera |
| US8009928B1 (en) * | 2008-01-23 | 2011-08-30 | A9.Com, Inc. | Method and system for detecting and recognizing text in images |
| CN103927534A (en) * | 2014-04-26 | 2014-07-16 | 无锡信捷电气股份有限公司 | Sprayed character online visual detection method based on convolutional neural network |
| CN104658011A (en) * | 2015-01-31 | 2015-05-27 | 北京理工大学 | Intelligent transportation moving object detection tracking method |
| US9405985B1 (en) * | 2015-03-17 | 2016-08-02 | Xerox Corporation | Leveraging character-by-character image classifiers to improve license plate state identification |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9354778B2 (en) * | 2013-12-06 | 2016-05-31 | Digimarc Corporation | Smartphone-based methods and systems |
-
2019
- 2019-01-28 CN CN201910077214.1A patent/CN109785319B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6553131B1 (en) * | 1999-09-15 | 2003-04-22 | Siemens Corporate Research, Inc. | License plate recognition with an intelligent camera |
| US8009928B1 (en) * | 2008-01-23 | 2011-08-30 | A9.Com, Inc. | Method and system for detecting and recognizing text in images |
| CN103927534A (en) * | 2014-04-26 | 2014-07-16 | 无锡信捷电气股份有限公司 | Sprayed character online visual detection method based on convolutional neural network |
| CN104658011A (en) * | 2015-01-31 | 2015-05-27 | 北京理工大学 | Intelligent transportation moving object detection tracking method |
| US9405985B1 (en) * | 2015-03-17 | 2016-08-02 | Xerox Corporation | Leveraging character-by-character image classifiers to improve license plate state identification |
Non-Patent Citations (1)
| Title |
|---|
| 车凯 ; 向郑涛 ; 陈宇峰 ; 吕坚 ; 周云.基于改进Fast R-CNN的红外图像行人检测研究.《红外技术》.2018,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109785319A (en) | 2019-05-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109785319B (en) | System, device and method for detecting code-spraying character based on image processing | |
| CN107563446B (en) | Target detection method for micro-operation system | |
| CN105894499B (en) | A kind of space object three-dimensional information rapid detection method based on binocular vision | |
| CN109308693A (en) | By the target detection and pose measurement list binocular vision system of a ptz camera building | |
| CN109859272B (en) | Automatic focusing binocular camera calibration method and device | |
| Semeniuta | Analysis of camera calibration with respect to measurement accuracy | |
| CN109986172B (en) | Welding seam positioning method, equipment and system | |
| CN110108712A (en) | Multifunctional visual sense defect detecting system | |
| CN109142509B (en) | Round steel magnetic powder flaw detection method and device | |
| CN109916914B (en) | Product defect detection method and device | |
| CN108827979A (en) | A kind of module group lens appearance detecting method | |
| CN110738644A (en) | automobile coating surface defect detection method and system based on deep learning | |
| CN113252567A (en) | Method, system, medium and terminal for rapidly detecting multiple defects on surface of aluminum valve plate | |
| CN109358067A (en) | Motor ring varistor defect detecting system based on computer vision and method | |
| CN113847881A (en) | Free-form surface profile tolerance detection method based on machine vision | |
| CN114359405A (en) | Calibration method of off-axis Samm 3D line laser camera | |
| CN108156359A (en) | Intelligent industrial camera | |
| CN115760893A (en) | Single droplet particle size and speed measuring method based on nuclear correlation filtering algorithm | |
| CN111833350B (en) | Machine vision detection method and system | |
| CN107248151B (en) | Intelligent liquid crystal display detection method and system based on machine vision | |
| Geng et al. | Machine vision detection method for surface defects of automobile stamping parts | |
| CN111486790A (en) | Full-size detection method and device for battery | |
| CN115908275A (en) | Hot ring rolling deformation geometric state online measurement method based on deep learning | |
| CN114755236A (en) | System and method for detecting surface defects of electroplated part with revolution curved surface | |
| Lei et al. | Automated vision inspection system for the size measurement of workpieces |
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 |