CN102890822A - Device with function of detecting object position, and detecting method of device - Google Patents
Device with function of detecting object position, and detecting method of device Download PDFInfo
- Publication number
- CN102890822A CN102890822A CN2011102035054A CN201110203505A CN102890822A CN 102890822 A CN102890822 A CN 102890822A CN 2011102035054 A CN2011102035054 A CN 2011102035054A CN 201110203505 A CN201110203505 A CN 201110203505A CN 102890822 A CN102890822 A CN 102890822A
- Authority
- CN
- China
- Prior art keywords
- under test
- object under
- module
- weight allocation
- 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.)
- Granted
Links
Images
Abstract
The invention discloses a device with a function of detecting an object position and a detecting method of the device. The device comprises an image acquisition module, a weight allocation module and a processing module; the image acquisition module acquires an image; the weight allocation module carries out weight allocation according to the image and prior information of a to-be-measured object and calculates an initial center of gravity of the to-be-measured object according to the possible initial position of the to-be-measured object; the processing module carries out statistics and analysis based on the result of the weight allocation and the initial center of gravity and obtains an analysis result for updating the possible position of the to-be-measured object; the processing module judges whether the analysis result meets a preset value, if so, an estimated result is generated; and if not, the steps are repeated.
Description
Technical field
The present invention relates to a kind of image detecting technology, espespecially a kind of image detecting technology of coming the particular location of particular artifact in the rapid detection image with iteration algorithm.
Background technology
In recent years, intelligent image detecting technology more and more comes into one's own, and it can be applied to many different electronic products, for example can see the application of intelligent image detecting technology on digital still camera, the interactive product of body sense and the safety monitoring system.And in these were used, the shape information of object was being played the part of very important role at image velocity detection and quality.A good shape information even be approximate shapes size (Approximated Shape), can both be accelerated significantly detecting speed and improve the detecting quality.
Generally speaking, estimate the shape information that particular artifact in the image, can utilize point by point (pixel-by-pixel) comparison of default color information, and enlarge gradually the scope of comparison, until the color value of pixel does not meet default condition in the image.In the field of Image Segmentation (Image Segmentation), these class methods often are used.For example, watershed method (WaterShed) and regional flop-in method (Region Growing) etc.Although these class methods can provide all well and good estimation effect on the simple object of color, it is on the computing velocity, then have many restrictions in the setting of color continuity (Connected Component) and Sharp criteria.
Yet in order to carry out to the object of color complexity the shape estimation, the method (Graph-Based Method) take graph theory as the basis provides effective solution in recent years.For example, Graph-Cut and Normalized cuts etc.Although above-mentioned method can accurately be estimated the shape of the object of color complexity, it also has very high complexity on calculating.Similarly, the method (Model-Based Method) take mould shape as the basis also has identical problem.For example, Active contour (Active Contour Method).Therefore, how to accelerate detecting speed, improve the detecting quality, solve the successional restriction of color, improve because Sharp criteria is difficult for defining and cause elasticity or fault-tolerance not enough and reduce the complexity of calculating when the object of color complexity estimated and be the problem that institute of the present invention wish solves.
Summary of the invention
Because the problem of above-mentioned known techniques, the object of the present invention is to provide a kind of device and method for detecting thereof with sensing object position function, detecting speed is slow to solve in the known techniques, detecting poor quality, elasticity or fault-tolerance is not enough, the color continuity is restricted and when the object of color complexity estimated, computation complexity is crossed the problems such as high, and the sensing object position function can be applied on the limited hand-held action device of computational resource, such as numerical digit camera, mobile phone etc.
To achieve these goals, the present invention has adopted following technical scheme:
According to purpose of the present invention, a kind of device with sensing object position function is proposed, it is characterized in that, comprising: an image capture module captures an image data; One weight allocation module, according to the image in one of the object under test in the image data possible initial position and object under test one in advance information (Priori Knowledge) carry out a weight allocation, and utilize possible initial position to calculate an original center of gravity of object under test; One processes module, carries out a statistical study according to result and original center of gravity that weight allocation produces, to calculate a possible position of object under test, and try to achieve barycentric coordinates, and whether the result who judges statistical study meet a preset value, if then produce an estimation result; If not, then replace possible position and the barycentric coordinates that possible initial position calculates object under test again with described possible position.
Wherein, weight allocation in advance information converts probability density function (Probability Density Function to, PDF), and according to the color value of each pixel in the image in the possible initial position in the image data (or edge direction values or its in conjunction with) and among probability density function respective value, respectively given each pixel one weighted value.
Wherein, statistical study utilizes weighted value, and calculate each pixel with respect to the spatial relationship of center of gravity, try to achieve a weighting covariation matrix of object under test, and weighting covariation matrix (Weighted Covariance Matrix) is carried out Eigenvalues Decomposition (Eigen Decomposition) or svd (Singular Value Decomposition).
Wherein, preset value is default degree of accuracy, default iterations or its combination.
Wherein, information comprises color distribution information, GTG brightness statistics figure (Histogram), edge direction gradient statistical graph (Edge Orientation/Gradient Histogram) or its combination of object under test in advance.
Wherein, in advance information by training data (Training Data), manually specify or grouping method (Clustering) is produced, and the possible initial position of object under test is produced by artificial appointment, grouping method or random sample consensus (Random Sample Consensus, RANSAC).
According to purpose of the present invention, reintroduce a kind of sensing object position method, be applicable to a device with sensing object position function, described device comprises an image capture module, a weight allocation module and a processing module.The sensing object position method comprises the following steps: to capture an image data with an image capture module; Utilize a weight allocation module according to the image in one of an object under test among the image data possible initial position and object under test one in advance information carry out a weight allocation, and calculate an original center of gravity of object under test according to one of object under test possible initial position; Process module by one and carry out a statistical study according to result and original center of gravity that weight allocation produces, calculating a possible position of object under test, and try to achieve barycentric coordinates; Replace possible initial position by processing module with possible position, and repeat above-mentioned steps, until the result of statistical study meets a preset value; Produce an estimation result by processing module.
According to purpose of the present invention, reintroduce a kind of device with sensing object position function, it is characterized in that, comprising: an image capture module captures an image data; One computing module, according to the image in one of the object under test in the described image data possible initial position and described object under test one in advance information carry out a weight allocation, and utilize described possible initial position to calculate an original center of gravity of described object under test, and result and described original center of gravity that the described weight allocation of foundation produces are carried out a statistical study, calculating a possible position of described object under test, and try to achieve barycentric coordinates; And one process module, judges whether the result of described statistical study meets a preset value, if then produce an estimation result; If not, then calculate again the described possible position of described object under test, and described barycentric coordinates.
According to purpose of the present invention, reintroduce a kind of device with sensing object position function, it is characterized in that, comprising: an image capture module captures an image data; One weight allocation module, according to the image in the original shape limitting casing of the object under test in the image data and object under test one in advance information carry out a weight allocation, and utilize the original shape limitting casing to calculate an original center of gravity of object under test; One computing module carries out a statistical study according to result and original center of gravity that weight allocation produces, calculating a shape limitting casing of object under test, and tries to achieve barycentric coordinates; One processes module, judges whether the result of statistical study meets a preset value, if then produce an estimation result; If not, then replace shape limitting casing and the barycentric coordinates that the original shape limitting casing calculates object under test again with the shape limitting casing.
Device and method for detecting thereof with sensing object position function of the present invention has following advantage:
(1) device with sensing object position function of the present invention and method for detecting thereof can be by obtaining in advance the in advance information quality accelerating to detect the speed of object space and improve the detecting object such as color, edge direction gradient statistical graph.
(2) device and the method for detecting thereof with sensing object position function of the present invention is the position of detecting object take statistical theory as the iteration algorithm on basis, and therefore, it can improve the problem of color continuity restriction.
(3) device with sensing object position function of the present invention and method for detecting thereof are to calculate more easily such as probability density function and weighting covariation matrix etc. by repeating, therefore, even color is estimated than complex object, do not have too large computation complexity yet.
(4) device and the method for detecting thereof with sensing object position function of the present invention is to utilize Sharp criteria to set easy iteration algorithm, and therefore, it has higher elasticity and fault-tolerance.
Description of drawings
Fig. 1 is the calcspar of the first embodiment of the device with sensing object position function of the present invention.
Fig. 2 is the operational flowchart of the first embodiment of the device with sensing object position function of the present invention.
Fig. 3 is the calcspar of the second embodiment of the device with sensing object position function of the present invention.
Fig. 4 is the schematic diagram of the second embodiment of the device with sensing object position function of the present invention.
Fig. 5 is the operational flowchart of the second embodiment of the device with sensing object position function of the present invention.
Fig. 6 is the process flow diagram of sensing object position method of the present invention.
Embodiment
See also Fig. 1, it is the calcspar of the first embodiment of the device with sensing object position function of the present invention.As shown in the figure, image capture unit 1 comprises an image capture module 11, a computing module 12 and a processing module 13.Wherein, computing module 12 is electrically connected image capture module 11 and processes module 13, processes module 13 and is electrically connected image capture module 11.
Wherein, image capture unit 1 can be the multiple electronic equipment that needs the definite position of detecting particular artifact in image.For example, safety monitoring system, home video safety system, digital still camera, man-machine interface (Human Computer Interface) and body sense interaction product etc.For ease of more accurately understanding technical characterictic of the present invention, below take digital still camera as embodiment, but not as limit.
The image data 111 of image capture module 11 acquisitions one object under test 14, its computing module 12 can be according to the in advance information (Priori Knowledge) of object under test 14, with each pixel 121 in the image 1111 in specific region in the image data 111 (being the possible initial position of object under test 14), do one by one weight allocation 126, and obtain its weighted value 122.And the possible initial position of object can appear at position in the image data 111 by artificial object under test 14 most probables of directly specifying, perhaps, with grouping method (Clustering) and random sample consensus (Random Sample Consensus, RANSAC) etc. algorithm is estimated, and the possible initial position of object has been arranged, just can calculate original center of gravity 123.
Will be in advance information integration to image capture unit 1 can accelerate significantly object detecting speed, and information can be color distribution information, GTG brightness statistics figure (Histogram), edge direction gradient statistical graph (Edge Orientation/Gradient Histogram) or its combination of object under test in advance.In addition, in the acquisition mode of information in advance, modal have a training data (Training Data), take apple as example, can use polytype apple data as the training data bank, with as the basis of detecting.In addition, can use simple grouping method (Clustering) or directly living by the finger prosthesis fixed output quota.
Aforesaid weighted value 122 and original center of gravity 123 have been arranged, and computing module 12 just can carry out statistical study 127, and its mode can be sets up a statistics model, and it is carried out numerical analysis, to calculate object under test possible position 124 and barycentric coordinates 125.Main shaft of the present invention is to take a kind of iteration algorithm (Iterative Method), carry out repeatedly weight allocation 126 and statistical study 127, to repeat to upgrade the possible position of object, until when satisfying the degree of accuracy of certain iterations or object detecting, process module 13 and namely produce an estimation result.
In known techniques, utilize the color information to compare point by point, and the method for the scope of pointwise expansion comparison, iteration algorithm of the present invention has concept and the characteristic of dynamic step number (Adaptive Step-Size), can significantly improve detecting efficient.
See also Fig. 2, it is the operational flowchart of the first embodiment of the device with sensing object position function of the present invention.
In step S21, image capture module pick-up image data.
In step S22, the computing module carries out weight allocation according to each pixel of the image in the possible initial position of object under test in the in advance information of object under test and the image data, to obtain the weighted value of each pixel.
In step S23, the computing module calculates the original center of gravity of object under test according to the possible initial position of object under test.
In step S24, the computing module carries out statistical study according to weighted value and original center of gravity, to calculate possible position and the barycentric coordinates of object under test.
In step S25, process module and judge that the result of statistical study meets preset value? if then enter step S26; If not, then get back to and enter step S251, process module and replace possible initial position with possible position, and indication computing module recomputates possible position and barycentric coordinates.
In step S26, process module and produce the estimation result.
See also Fig. 3, it is the calcspar of the second embodiment of the device with sensing object position function of the present invention.As shown in the figure, image capture unit 3 comprises an image capture module 31, a weight allocation module 32 and a processing module 33.Wherein, weight allocation module 32 is electrically connected image capture module 31 and processes module 33; Process module 33 and be electrically connected image capture module 31.
At first, image capture module 31 acquisitions one image data 311.And weight allocation module 32 can convert the in advance information of object under test 34 to probability density function 324 (Probability Density Function, PDF), and according to color value, edge direction values and the respective value in probability density function 324 thereof of each pixel 321 in the image 3111 in the possible initial position of object under test 34 in the image data 311, given its weighted value 322 respectively.And the possible initial position of object can be estimated by aforesaid method, and the possible initial position of object has been arranged, and just can calculate original center of gravity 323.
At this moment, process 33 of modules and utilize above-mentioned weighted value 322, and calculate each pixel 321 with respect to the spatial relationship of center of gravity, try to achieve the weighting covariation matrix 331 of object under test 34, and weighting covariation matrix 331 is carried out Eigenvalues Decomposition (Eigen Decomposition) or svd 332 (Singular Value Decomposition).Wherein, the eigenwert of trying to achieve (Eigen Value) is major axis and the minor axis of the shape limitting casing of object under test 34, and the anglec of rotation of object under test 34 also can be tried to achieve by eigenwert.And the major axis that utilization is tried to achieve and the possible position 334 (shape limitting casing) of the renewable object under test of minor axis, and again try to achieve barycentric coordinates 335, to carry out iterative computation again.When the possible position 334 of the object under test 34 of trying to achieve meets default degree of accuracy, or iterations is when having reached a preset times, processes 33 of modules and can produce the estimation result, uses so that other handling procedures or application program to be provided.For example, the estimation result can comprise the possible position 334 that calculates at last, provides body form identification handling procedure, object classification identification handling procedure to use, and can dwindle by this imagery zone that needs processing; Or according to the possible position 334 that calculates at last, illustrate a picture frame at the display that installs, watch for the user.
It is worth mentioning that, iteration algorithm used in the present invention is as the basis take statistical theories such as probability density functions.That is to say, be the method on basis compared to the watershed method of known techniques and regional flop-in method etc. take pixel, method used in the present invention can be improved the problem of color continuity restriction.In addition, the end condition of iteration calculation is set also very simple and easy, and therefore, comparing with known techniques has better elasticity and fault-tolerance.
In addition, in technical field of the present invention, the general knowledge person can combine each functional module arbitrarily and become a device of integrating, and also each functional module separately can be formed the device that has more refinement function for a plurality of.
See also Fig. 4, it is the schematic diagram of the second embodiment of the device with sensing object position function of the present invention.As shown in the figure, user's operand bit camera is taken an image, and the default position of original shape limitting casing 41 (being the possible initial position of apple) in the middle of camera lens of apple.At this moment, the weight allocation module is analyzed each pixel in the image in the original shape limitting casing 41, its in advance information with apple is converted to probability density function, again with each pixel according to probability density function, distinguish given its weighted value, and calculate original center of gravity according to original shape limitting casing 41.The computing module then carries out the calculating of aforesaid weighting covariation matrix, the more position of new shape limitting casing.As can be seen from Figure, process is eight times iterative computation approximately, and the result of detecting is quite accurate.
Do not need to set up in advance the object feature data of wish detecting in the device of the present invention, so be not limited to the application of digital still camera, it can be applied to the various systems that need detecting particular artifact position.For example, home video safety system etc., even can be applied in the traffic safety surveillance, whether be used for monitoring has the unusual conditions such as rule-breaking vehicle parking.
See also Fig. 5, it is the operational flowchart of the second embodiment of the device with sensing object position function of the present invention.
In step S51, image capture module pick-up image data.
In step S52, the weight allocation module converts the in advance information of object under test to probability density function, and according to color value, edge direction values and the respective value in probability density function thereof of each pixel in the image in the possible initial position of object under test in the image data, given its weighted value respectively.
In step S53, the weight allocation module calculates the original center of gravity of object under test according to the possible initial position of object under test.
In step S54, the computing module is according to weighted value, and calculates each pixel with respect to the spatial relationship of center of gravity, tries to achieve the weighting covariation matrix of object under test, and the weighting covariation matrix carried out Eigenvalues Decomposition or svd, to calculate possible position and the barycentric coordinates of object under test.
In step S55, process module and judge that possible position and the barycentric coordinates of object under test meet preset value? if then enter step S56; If not, then enter step S551, process module and replace possible initial position with possible position, and indication computing module recomputates possible position and barycentric coordinates.
In step S56, process module and produce the estimation result.
See also Fig. 6, it is the process flow diagram of object detecting method of the present invention.As shown in the figure, object detecting method of the present invention is applicable to a device with sensing object position function, and described device comprises an image capture module, a weight allocation module, a computing module and a processing module.The object detecting method comprises the following steps:
(S61) with image capture module pick-up image data.
(S62) utilize the weight allocation module to carry out weight allocation according to the in advance information of the image in the possible initial position of object under test among the image data and object under test, and calculate the original center of gravity of object under test according to the possible initial position of object under test.
(S63) carry out statistical study by processing module according to result and original center of gravity that weight allocation produces, calculating the possible position of object under test, and try to achieve barycentric coordinates.
(S64) the possible initial position by the object under test in the possible position step of replacing (S62) of processing the object under test that module gets with calculating, and repeating step (S62) and (S63), until the result of statistical study meets preset value.
(S65) produce an estimation result by processing module.
Described when the detailed description of object detecting method of the present invention and embodiment are narrated the device with sensing object position function of the present invention in front, at this for schematic illustration, just repeated description no longer.
In sum, device with sensing object position function of the present invention and method for detecting thereof utilize in advance information and take the iteration algorithm of statistical theory as the basis, particular artifact in the image capture unit is carried out the estimation of shape and position, therefore, can improve the problem of color continuity restriction, on detecting speed and quality, also have greatly improved.In addition, iteration algorithm of the present invention is to calculate more easily such as probability density function and weighting covariation matrix etc. by repeating, therefore, compare with known techniques, complexity is lower in the calculating, and the setting of Sharp criteria is also relatively easy, for example, can set end condition with the number of times of iteration, therefore, be fit to very much to be applied in intelligent image detecting technical.
The above is preferred embodiment of the present invention and the know-why used thereof; for a person skilled in the art; in the situation that do not deviate from the spirit and scope of the present invention; any based on apparent changes such as the equivalent transformation on the technical solution of the present invention basis, simple replacements, all belong within the protection domain of the present invention.
Claims (16)
1. the device with sensing object position function is characterized in that, comprising:
One image capture module captures an image data;
One weight allocation module, according to the image in one of the object under test in the described image data possible initial position and described object under test one in advance information carry out a weight allocation, and utilize described possible initial position to calculate an original center of gravity of described object under test; And
One processes module, carry out a statistical study according to result and described original center of gravity that described weight allocation produces, to calculate a possible position of described object under test, and try to achieve barycentric coordinates, and whether the result who judges described statistical study meets a preset value, if then produce an estimation result; If not, then replace described possible position and the described barycentric coordinates that described possible initial position calculates described object under test again with described possible position.
2. the device with sensing object position function as claimed in claim 1, it is characterized in that: described weight allocation converts described in advance information to probability density function, and according to color value, edge direction values and the respective value among probability density function thereof of each pixel in the described image in the described possible initial position in the described image data, respectively given described each pixel one weighted value.
3. the device with sensing object position function as claimed in claim 2, it is characterized in that: described statistical study utilizes described weighted value, and calculate described each pixel with respect to the spatial relationship of center of gravity, try to achieve a weighting covariation matrix of described object under test, and described weighting covariation matrix is carried out Eigenvalues Decomposition or svd.
4. the device with sensing object position function as claimed in claim 1 is characterized in that: described preset value is default degree of accuracy or default iterations or its combination.
5. the device with sensing object position function as claimed in claim 1, it is characterized in that: described in advance information comprises color distribution information or GTG brightness statistics figure or edge direction gradient statistical graph or its combination of described object under test.
6. the device with sensing object position function as claimed in claim 1, it is characterized in that: described in advance information is by the training data or manually specify or grouping method is produced, and the described possible initial position of described object under test is produced by artificial appointment or grouping method or random sample consensus.
7. a sensing object position method is applicable to a device with sensing object position function, it is characterized in that, comprises the following steps:
Capture an image data with an image capture module;
Utilize a weight allocation module according to the image in one of an object under test among the described image data possible initial position and described object under test one in advance information carry out a weight allocation, and calculate an original center of gravity of described object under test according to described possible initial position;
Process module by one and carry out a statistical study according to result and described original center of gravity that described weight allocation produces, calculating a possible position of described object under test, and try to achieve barycentric coordinates;
Replace described possible initial position by described processing module with described possible position, and repeat above-mentioned steps, until the result of described statistical study meets a preset value; And
Produce an estimation result by processing module.
8. sensing object position method as claimed in claim 7, it is characterized in that: described weight allocation converts described in advance information to probability density function, and according to color value, edge direction values and respective value, respectively given described each pixel one weighted value among described probability density function of each pixel in the described image in initial position possible described in the described image data.
9. sensing object position method as claimed in claim 8, it is characterized in that: described statistical study utilizes described weighted value, and calculate described each pixel with respect to the spatial relationship of center of gravity, try to achieve a weighting covariation matrix of described object under test, and described weighting covariation matrix is carried out Eigenvalues Decomposition or svd.
10. sensing object position method as claimed in claim 7 is characterized in that: described preset value is default degree of accuracy or default iterations or its combination.
11. sensing object position method as claimed in claim 7 is characterized in that: described in advance information comprises color distribution information or GTG brightness statistics figure or edge direction gradient statistical graph or its combination of described object under test.
12. sensing object position method as claimed in claim 7, it is characterized in that: described in advance information is by the training data or manually specify or grouping method is produced, and the described possible initial position of described object under test is produced by artificial appointment or grouping method or random sample consensus.
13. the device with sensing object position function is characterized in that, comprising:
One image capture module captures an image data;
One weight allocation module, according to the image in the original shape limitting casing of the object under test in the described image data and described object under test one in advance information carry out a weight allocation, and utilize described original shape limitting casing to calculate an original center of gravity of described object under test;
One computing module carries out a statistical study according to result and described original center of gravity that described weight allocation produces, calculating a shape limitting casing of described object under test, and tries to achieve barycentric coordinates; And
One processes module, judges whether the result of described statistical study meets a preset value, if then produce an estimation result; If not, then replace described shape limitting casing and the described barycentric coordinates that described original shape limitting casing calculates described object under test again with described shape limitting casing.
14. the device with sensing object position function as claimed in claim 13, it is characterized in that: described weight allocation converts described in advance information to probability density function, and according to color value, edge direction values and respective value, respectively given described each pixel one weighted value among described probability density function of each pixel in the described image in described original shape limitting casing.
15. the device with sensing object position function as claimed in claim 14, it is characterized in that: described statistical study utilizes described weighted value, and calculate described each pixel with respect to the spatial relationship of center of gravity, try to achieve a weighting covariation matrix of described object under test, and described weighting covariation matrix is carried out Eigenvalues Decomposition or svd.
16. the device with sensing object position function as claimed in claim 13 is characterized in that: described preset value is default degree of accuracy or default iterations or its combination.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110203505.4A CN102890822B (en) | 2011-07-20 | 2011-07-20 | Device with function of detecting object position, and detecting method of device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110203505.4A CN102890822B (en) | 2011-07-20 | 2011-07-20 | Device with function of detecting object position, and detecting method of device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102890822A true CN102890822A (en) | 2013-01-23 |
| CN102890822B CN102890822B (en) | 2015-05-20 |
Family
ID=47534316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110203505.4A Expired - Fee Related CN102890822B (en) | 2011-07-20 | 2011-07-20 | Device with function of detecting object position, and detecting method of device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102890822B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104731310A (en) * | 2013-12-23 | 2015-06-24 | 原相科技股份有限公司 | Optical object tracking method and optical object tracking system |
| CN106997600A (en) * | 2016-01-22 | 2017-08-01 | 宏达国际电子股份有限公司 | Motion detection device and repetitive operation method for detecting |
| TWI606886B (en) * | 2016-11-15 | 2017-12-01 | 北鉅精機股份有限公司 | Intellectual system for atc tool exchanging speed moderation |
| CN113275929A (en) * | 2021-03-25 | 2021-08-20 | 昆山北钜机械有限公司 | Automatic light and heavy cutter distinguishing device and method for cutter head of cutter library |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0835814A (en) * | 1994-07-22 | 1996-02-06 | Japan Steel Works Ltd:The | Method and apparatus for obtaining target position |
| CN1619593A (en) * | 2004-12-09 | 2005-05-25 | 上海交通大学 | Video frequency motion target adaptive tracking method based on multicharacteristic information fusion |
-
2011
- 2011-07-20 CN CN201110203505.4A patent/CN102890822B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0835814A (en) * | 1994-07-22 | 1996-02-06 | Japan Steel Works Ltd:The | Method and apparatus for obtaining target position |
| CN1619593A (en) * | 2004-12-09 | 2005-05-25 | 上海交通大学 | Video frequency motion target adaptive tracking method based on multicharacteristic information fusion |
Non-Patent Citations (4)
| Title |
|---|
| 吴波 等: "一种稳健的基于主颜色的多目标跟踪算法", 《中国科学院研究生院学报》 * |
| 王俊骅 等: "基于OpenCV和Halcon的交通冲突视频自动检测及数据处理", 《同济大学学报（自然科学版）》 * |
| 王文成 等: "一种改进的重心法在光带中心提取中的应用", 《电子●激光》 * |
| 胡峰俊 等: "交通流视频检测中车辆的检测与跟踪", 《计算机工程与设计》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104731310A (en) * | 2013-12-23 | 2015-06-24 | 原相科技股份有限公司 | Optical object tracking method and optical object tracking system |
| CN104731310B (en) * | 2013-12-23 | 2018-03-30 | 原相科技股份有限公司 | Optical profile type object tracking method and its related optical formula tracing system |
| CN106997600A (en) * | 2016-01-22 | 2017-08-01 | 宏达国际电子股份有限公司 | Motion detection device and repetitive operation method for detecting |
| CN106997600B (en) * | 2016-01-22 | 2019-12-13 | 宏达国际电子股份有限公司 | Motion detection device and repetitive motion detection method |
| TWI606886B (en) * | 2016-11-15 | 2017-12-01 | 北鉅精機股份有限公司 | Intellectual system for atc tool exchanging speed moderation |
| CN113275929A (en) * | 2021-03-25 | 2021-08-20 | 昆山北钜机械有限公司 | Automatic light and heavy cutter distinguishing device and method for cutter head of cutter library |
| CN113275929B (en) * | 2021-03-25 | 2022-08-30 | 昆山北钜机械有限公司 | Automatic light and heavy cutter distinguishing device and method for cutter head of cutter library |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102890822B (en) | 2015-05-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhao et al. | TBC-Net: A real-time detector for infrared small target detection using semantic constraint | |
| Gao et al. | A novel visual tracking method using bat algorithm | |
| US8792722B2 (en) | Hand gesture detection | |
| CN102982341B (en) | Self-intended crowd density estimation method for camera capable of straddling | |
| Kurata et al. | The hand mouse: GMM hand-color classification and mean shift tracking | |
| TWI459307B (en) | Apparatus for detecting object in image and detecting method thereof | |
| WO2018090937A1 (en) | Image processing method, terminal and storage medium | |
| CN102982537B (en) | A kind of method and system detecting scene change | |
| CN107301376B (en) | Pedestrian detection method based on deep learning multi-layer stimulation | |
| CN107944381B (en) | Face tracking method, face tracking device, terminal and storage medium | |
| CN106648078A (en) | Multimode interaction method and system applied to intelligent robot | |
| CN110135369A (en) | A kind of Activity recognition method, system, equipment and computer readable storage medium | |
| CN102890822B (en) | Device with function of detecting object position, and detecting method of device | |
| CN113673465A (en) | Image detection method, device, equipment and readable storage medium | |
| CN109711287A (en) | Face acquisition method and Related product | |
| CN105118073A (en) | Human body head target identification method based on Xtion camera | |
| CN105787963B (en) | A kind of video target tracking method and device | |
| CN111368596B (en) | Face recognition backlight compensation method and device, readable storage medium and equipment | |
| CN112700568A (en) | Identity authentication method, equipment and computer readable storage medium | |
| CN116619780A (en) | Intelligent production method and system of phenolic composite material | |
| Zhang et al. | High-speed vision extraction based on the CamShift algorithm | |
| CN115601834A (en) | Fall detection method based on WiFi channel state information | |
| CN104680122A (en) | Tracking method and device based on skin color detection | |
| Sun et al. | Research on the embedded system of facial expression recognition based on HMM | |
| Song et al. | Behavior Recognition of the Elderly in Indoor Environment Based on Feature Fusion of Wi-Fi Perception and Videos |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150520 Termination date: 20190720 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |