CN111191550B - Visual perception device and method based on automatic dynamic adjustment of image sharpness - Google Patents
Visual perception device and method based on automatic dynamic adjustment of image sharpness Download PDFInfo
- Publication number
- CN111191550B CN111191550B CN201911338461.9A CN201911338461A CN111191550B CN 111191550 B CN111191550 B CN 111191550B CN 201911338461 A CN201911338461 A CN 201911338461A CN 111191550 B CN111191550 B CN 111191550B
- Authority
- CN
- China
- Prior art keywords
- picture
- result
- sharpness
- sharpening
- probability
- 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
- 230000016776 visual perception Effects 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 title claims abstract description 15
- 238000000605 extraction Methods 0.000 claims abstract description 10
- 238000004458 analytical method Methods 0.000 claims abstract description 8
- 238000012163 sequencing technique Methods 0.000 claims abstract description 4
- 125000004122 cyclic group Chemical group 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 4
- 230000000007 visual effect Effects 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 3
- 230000004438 eyesight Effects 0.000 description 6
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003707 image sharpening Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 210000000653 nervous system Anatomy 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G06T5/73—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/49—Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
Abstract
The invention provides a visual perception device and a visual perception method based on automatic dynamic adjustment of image sharpness, which are characterized in that dynamic video stream data are acquired through a high-definition camera in real time and dynamic video, and key frame extraction is carried out on the dynamic video stream data at high frequency; sequencing the extracted pictures of the key frames according to the time sequence to form a picture sequence which can be gradually overturned and reflected before and after; sharpening the key frame pictures to generate pictures conforming to the appointed sharpening degree, carrying out depth recognition analysis on each generated picture, and constructing a set of recognition results of the single picture; and calculating probability distribution of the result set to obtain a result sample with the highest probability as a visual perception result of the key frame picture. The invention utilizes the automatic dynamic adjustment of the image sharpness to improve the accuracy of computer identification.
Description
Technical Field
The invention belongs to the technical field of visual perception, and particularly relates to a visual perception device and method based on automatic dynamic adjustment of image sharpness.
Background
In recent years, with the rise of artificial intelligence and robotics, mankind has been dreamed to be able to have a robot system equipped with artificial intelligence vision technology possess a human-like vision system that can "see" and perceive something of interest in front of it. The problem of having a machine with human general vision is particularly sophisticated and complex because the current intelligent robotic systems are not able to simulate the human visual nervous system and consciousness. How to make accurate judgment on video and image signals captured by a camera of an intelligent robot system and identify interesting useful information of corresponding things becomes one of key problems facing the field of computer vision identification.
Disclosure of Invention
The invention aims to provide a visual perception device and a visual perception method based on automatic dynamic adjustment of image sharpness, which utilize the automatic dynamic adjustment of the image sharpness to improve the accuracy of computer identification.
The invention provides the following technical scheme:
a visual perception device based on automatic dynamic adjustment of image sharpness comprises a video acquisition processing device and a visual perception device which are connected with each other;
the video acquisition and processing device comprises a video acquisition and extraction device and a key frame segmentation device, wherein the video acquisition and extraction device acquires dynamic video stream data in real time and extracts key frames at a high frequency, and the key frame segmentation device sorts the extracted pictures of the key frames according to time sequence to form a picture sequence which can be progressively overturned and reflected before and after;
the visual sensing device comprises a picture sharpening device, an image sharpness dynamic adjustment device and an image recognition device, wherein the picture sharpening device sharpens a key frame picture to generate pictures conforming to a specified sharpening degree, the image sharpness dynamic adjustment device carries out depth recognition analysis on each picture generated by the picture sharpening device to construct a set of recognition results of a single picture, and the image recognition device calculates probability distribution of the set of results to take a result sample with the highest probability as a visual perception result of the key frame picture.
Preferably, the image sharpness dynamic adjustment device performs recognition and information extraction by performing a loop variable algorithm on each sharpness change of the picture to construct a recognition result and set of the single picture.
Preferably, the loop variable algorithm comprises the following steps: recording the current cycle number as i and the initial value as 1, and then:
a. if i=1, identifying the picture for the first time, and sharpening the picture according to default sharpness parameters;
b. identifying and extracting information of the sharpened picture, and recording the obtained information as a sample i of a result set space;
c. reading a preset identification frequency parameter;
d. judging whether the recognition times are reached, if so, exiting the result set construction, and if not, continuing to execute the next step;
e. let the cyclic variable i=i+1;
f. reading a pre-configured sharpness change step size parameter;
g. calculating a new round of identified sharpness parameters, the new round of identified sharpness parameters being equal to the previous round of identified sharpness parameters plus sharpness change step size parameters;
h. sharpening the picture according to the sharpening parameters obtained in the previous step
i. B, circulating to the step b, identifying and extracting information of the sharpened picture, and recording the obtained information as a sample i of a result set space;
when the cycle is finished, a set of recognition results of the picture is constructed.
Preferably, the image recognition device judges the result of the maximum probability, and when the probability is greater than 50%, the current sample of the result of the maximum probability is used as the visual perception result of the key frame picture; when the probability is less than or equal to 50%, the dynamic image sharpness adjusting device changes the step length change direction, a row result set is constructed in the new step length change direction, whether the probability of the result with the highest probability in the result set in the reduction direction exceeds 50% is judged, if so, the step length change direction is credible, and the image recognition device is transferred to continue to judge, otherwise, the result set in the increase direction and the result set in the reduction direction are combined, and a result sample with the highest probability is obtained.
A visual perception method based on automatic dynamic adjustment of image sharpness, comprising the steps of:
s1: acquiring dynamic video stream data through a high-definition camera real-time dynamic video, and extracting key frames of the dynamic video stream data at high frequency;
s2: sequencing the extracted pictures of the key frames according to the time sequence to form a picture sequence which can be gradually overturned and reflected before and after;
s3: sharpening the key frame pictures to generate pictures conforming to the appointed sharpening degree, carrying out depth recognition analysis on each generated picture, and constructing a set of recognition results of the single picture;
s4: and calculating probability distribution of the result set to obtain a result sample with the highest probability as a visual perception result of the key frame picture.
Preferably, the following loop is executed for identifying and extracting information for each sharpness change of the picture in S3: recording the current cycle number as i and the initial value as 1, and then:
a. if i=1, identifying the picture for the first time, and sharpening the picture according to default sharpness parameters;
b. identifying and extracting information of the sharpened picture, and recording the obtained information as a sample i of a result set space;
c. reading a preset identification frequency parameter;
d. judging whether the recognition times are reached, if so, exiting the result set construction, and if not, continuing to execute the next step;
e. let the cyclic variable i=i+1;
f. reading a pre-configured sharpness change step size parameter;
g. calculating a new round of identified sharpness parameters, the new round of identified sharpness parameters being equal to the previous round of identified sharpness parameters plus sharpness change step size parameters;
h. sharpening the picture according to the sharpening parameters obtained in the previous step
i. B, circulating to the step b, identifying and extracting information of the sharpened picture, and recording the obtained information as a sample i of a result set space;
when the cycle is finished, a set of recognition results of the picture is constructed.
Preferably, the method further comprises the following steps:
s5: judging whether the probability of the result of the maximum probability in the S4 exceeds 50%, and when the probability is more than 50%, taking the current sample of the result of the maximum probability as the visual perception result of the key frame picture;
when the probability is less than or equal to 50%, changing the sharpening step change direction in S3, constructing a row result set in the new step change direction, judging whether the probability of the result with the highest probability in the result set in the reduction direction exceeds 50%, if so, indicating that the step change direction is credible, and turning to the image recognition device to continue judging, otherwise, combining the result set in the increase direction and the result set in the reduction direction, and obtaining a result sample with the highest probability.
The beneficial effects of the invention are as follows: by applying the method and the device provided by the invention, on the basis of the traditional computer vision recognition method, the picture under each sharpness is recognized by cutting the video key frames and dynamically adjusting the sharpness of the picture, and reasonable analysis is carried out on the result set, so that the effect of computer vision recognition is obviously improved, and the application of related industries is helped to obtain a larger breakthrough.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a schematic view of the structure of the device of the present invention;
FIG. 2 is a schematic of the overall flow of the present invention;
FIG. 3 is a schematic diagram of an automatic dynamic adjustment flow of image sharpness according to the present invention;
fig. 4 is a schematic diagram of the sharpening direction backtracking flow of the present invention.
Detailed Description
As shown in fig. 1 and fig. 2, a visual perception device and a method based on automatic dynamic adjustment of image sharpness, the specific method is as follows:
1) And (3) real-time dynamic video acquisition:
the device comprises a real-time high-speed dynamic video acquisition module, wherein dynamic video stream data are acquired in real time through a high-speed high-definition camera;
2) Key frame high frequency extraction:
the key frame extraction is carried out on the dynamic video stream data acquired in the first step at a higher frequency, which is equivalent to cutting the video into a plurality of pictures very fast;
3) Arranging key frame pictures in time sequence:
sequencing the pictures cut in the second step according to the time sequence, so as to ensure that the pictures of the specific scene are adjacent to each other, and forming a picture sequence which can be gradually transited and reflected before and after;
4) Image sharpening algorithm:
the image sharpening algorithm subsystem provided by the invention can be used for sharpening an input original image according to different sharpening parameters to generate an image conforming to the appointed sharpening degree;
5) Constructing an identification result set of a single picture:
carrying out depth recognition analysis on each picture generated in the step 3;
specifically, as shown in fig. 3, for each sharpness change of a picture, the following loop is performed for identification and information extraction:
recording the current cycle number as i and the initial value as 1, and then:
a) If i=1, the identification is performed for the first time, and the picture is sharpened according to default sharpness parameters (which can be configured according to visual scene);
b) Identifying and extracting information of the sharpened picture, and recording the obtained information as a sample i of a result set space;
c) Reading a preset identification frequency parameter, and marking the identification frequency parameter as N;
d) Judging whether the recognition times are reached, if so, exiting the result set construction, and if not, continuing to execute the next step (e);
e) Let the cyclic variable i=i+1;
f) Reading a pre-configured sharpness change Step size parameter, and marking the Step;
g) Calculating a new round of recognized sharpness parameter, wherein the new round of recognized sharpness parameter=the previous round of recognized sharpness parameter+sharpness change Step length parameter, namely acutence=preauthance+step;
h) Sharpening the picture according to the sharpening parameters obtained in the previous step;
i) The step (b) is circulated, the sharpened picture is identified and information is extracted, and the obtained information is recorded as a sample i of a result set space;
when the cycle is finished, we construct a set of recognition results for the picture.
6) Sample result set probability distribution analysis:
6.1, calculating probability distribution of the result set;
and 6.2, taking the result with the highest occurrence probability as the identification result of the picture.
For example, if the recognition result set of a certain picture in a certain embodiment includes 10 samples as follows:
{“Result1” :“56339989”,
“Result2” :“56339999”,
“Result3” :“56339999”,
“Result4” :“66339999”,
“Result5” :“56339999”,
“Result6” :“66339999”,
“Result7” :“56339999”,
“Result8” :“56339999”,
“Result9” :“56339999”,
“Result10” :“56339999” }
the most frequent result is 56339999, with a probability of 70% in sample space, 66339999 next, 20% in sample space, 56339989 next, and 10% in sample space.
The distribution of the sample result set is obtained as follows:
7) Backtracking the sharpening direction:
7.1, judging whether the probability of the maximum probability in the last step exceeds 50 percent. If the step length is more than the step length, the step length change direction (increase) is credible, and the step length is transferred to the step 8 to continue the judgment;
otherwise, returning to the step 5, changing the step length change direction to be reduced, namely: acutance = preatance-Step;
7.2, continuing to construct a new result set in the new step change direction;
7.3, judging whether the probability of the result of the maximum probability in the result set in the decreasing direction exceeds 50%. If the step length is more than the step length, the step length change direction (reduction) is credible, and the step length is transferred to the step 8 to continue the judgment; otherwise, combining the result set in the increasing direction and the result set in the decreasing direction to obtain a result sample with the highest probability;
8) Judging visual perception results:
and taking the result sample with the highest probability in the previous step as the visual perception result of the key frame picture. For example, in the case of the sixth embodiment, it is finally determined that the information identified by the picture is 56339999.
The foregoing description is only a preferred embodiment of the present invention, and the present invention is not limited thereto, but it is to be understood that modifications and equivalents of some of the technical features described in the foregoing embodiments may be made by those skilled in the art, although the present invention has been described in detail with reference to the foregoing embodiments. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. The visual perception device based on the automatic dynamic adjustment of the image sharpness is characterized by comprising a video acquisition processing device and a visual perception device which are connected with each other;
the video acquisition and processing device comprises a video acquisition and extraction device and a key frame segmentation device, wherein the video acquisition and extraction device acquires dynamic video stream data in real time and extracts key frames at a high frequency, and the key frame segmentation device sorts the extracted pictures of the key frames according to time sequence to form a picture sequence which can be progressively overturned and reflected before and after;
the visual sensing device comprises a picture sharpening device, an image sharpness dynamic adjustment device and an image recognition device, wherein the picture sharpening device sharpens a key frame picture according to different sharpening parameters to generate pictures conforming to a specified sharpening degree, the image sharpness dynamic adjustment device carries out depth recognition analysis on each picture generated by the picture sharpening device to construct a set of recognition results of a single picture, and the image recognition device calculates probability distribution of the set of results to obtain a result sample with the highest probability as a visual perception result of the key frame picture;
the image sharpness dynamic adjustment device performs recognition and information extraction on each sharpness change of the picture by executing a circulation variable algorithm to construct a recognition result and a set of a single picture.
2. The visual perception apparatus based on automatic dynamic adjustment of image sharpness according to claim 1, wherein the loop variable algorithm comprises the steps of: recording the current cycle number as i and the initial value as 1, and then:
a. if i=1, identifying the picture for the first time, and sharpening the picture according to default sharpness parameters;
b. identifying and extracting information of the sharpened picture, and recording the obtained information as a sample i of a result set space;
c. reading a preset identification frequency parameter;
d. judging whether the recognition times are reached, if so, exiting the result set construction, and if not, continuing to execute the next step;
e. let the cyclic variable i=i+1;
f. reading a pre-configured sharpness change step size parameter;
g. calculating a new round of identified sharpness parameters, the new round of identified sharpness parameters being equal to the previous round of identified sharpness parameters plus sharpness change step size parameters;
h. sharpening the picture according to the sharpening parameters obtained in the previous step
i. B, circulating to the step b, identifying and extracting information of the sharpened picture, and recording the obtained information as a sample i of a result set space;
when the cycle is finished, a set of recognition results of the picture is constructed.
3. The visual perception device based on automatic dynamic adjustment of image sharpness according to claim 1, wherein the image recognition device judges the result of maximum probability, and when the probability is more than 50%, the current sample of the result of maximum probability is used as the visual perception result of the key frame picture; when the probability is less than or equal to 50%, the dynamic image sharpness adjusting device changes the step length change direction, a row result set is constructed in the new step length change direction, whether the probability of the result with the highest probability in the result set in the reduction direction exceeds 50% is judged, if so, the step length change direction is credible, and the image recognition device is transferred to continue to judge, otherwise, the result set in the increase direction and the result set in the reduction direction are combined, and a result sample with the highest probability is obtained.
4. A visual perception method based on automatic dynamic adjustment of image sharpness, comprising the steps of:
s1: acquiring dynamic video stream data through a high-definition camera real-time dynamic video, and extracting key frames of the dynamic video stream data at high frequency;
s2: sequencing the extracted pictures of the key frames according to the time sequence to form a picture sequence which can be gradually overturned and reflected before and after;
s3: sharpening the key frame pictures according to different sharpening parameters to generate pictures conforming to the appointed sharpening degree, carrying out depth recognition analysis on each generated picture, and constructing a set of recognition results of the single picture; the method comprises the steps of executing a cyclic variable algorithm to identify and extract information through each sharpness change of a picture, and constructing an identification result and a set of a single picture;
s4: and calculating probability distribution of the result set to obtain a result sample with the highest probability as a visual perception result of the key frame picture.
5. The visual perception method based on automatic dynamic adjustment of image sharpness according to claim 4, wherein the following loop is performed for identifying and extracting information for each sharpness change of a picture in S3: recording the current cycle number as i and the initial value as 1, and then:
a. if i=1, identifying the picture for the first time, and sharpening the picture according to default sharpness parameters;
b. identifying and extracting information of the sharpened picture, and recording the obtained information as a sample i of a result set space;
c. reading a preset identification frequency parameter;
d. judging whether the recognition times are reached, if so, exiting the result set construction, and if not, continuing to execute the next step;
e. let the cyclic variable i=i+1;
f. reading a pre-configured sharpness change step size parameter;
g. calculating a new round of identified sharpness parameters, the new round of identified sharpness parameters being equal to the previous round of identified sharpness parameters plus sharpness change step size parameters;
h. sharpening the picture according to the sharpening parameters obtained in the previous step
i. B, circulating to the step b, identifying and extracting information of the sharpened picture, and recording the obtained information as a sample i of a result set space;
when the cycle is finished, a set of recognition results of the picture is constructed.
6. The method for visual perception based on automatic dynamic adjustment of image sharpness according to claim 4, further comprising the steps of:
s5: judging whether the probability of the result of the maximum probability in the S4 exceeds 50%, and when the probability is more than 50%, taking the current sample of the result of the maximum probability as the visual perception result of the key frame picture;
when the probability is less than or equal to 50%, changing the sharpening step change direction in S3, constructing a row result set in the new step change direction, judging whether the probability of the result with the highest probability in the result set in the reduction direction exceeds 50%, if so, indicating that the step change direction is credible, and turning to the image recognition device to continue judging, otherwise, combining the result set in the increase direction and the result set in the reduction direction, and obtaining a result sample with the highest probability.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911338461.9A CN111191550B (en) | 2019-12-23 | 2019-12-23 | Visual perception device and method based on automatic dynamic adjustment of image sharpness |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911338461.9A CN111191550B (en) | 2019-12-23 | 2019-12-23 | Visual perception device and method based on automatic dynamic adjustment of image sharpness |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111191550A CN111191550A (en) | 2020-05-22 |
CN111191550B true CN111191550B (en) | 2023-05-02 |
Family
ID=70709338
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911338461.9A Active CN111191550B (en) | 2019-12-23 | 2019-12-23 | Visual perception device and method based on automatic dynamic adjustment of image sharpness |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111191550B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101996406A (en) * | 2010-11-03 | 2011-03-30 | 中国科学院光电技术研究所 | No-reference structural sharpness image quality evaluation method |
CN103279328A (en) * | 2013-04-08 | 2013-09-04 | 河海大学 | BlogRank algorithm parallelization processing construction method based on Haloop |
EP2696584A2 (en) * | 2012-08-06 | 2014-02-12 | Ricoh Company, Ltd. | Image capturing device and image capturing method |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103310435B (en) * | 2012-03-21 | 2016-04-06 | 华中科技大学 | Combine vertical projection and optimal path the method split characters on license plate |
CN103377468A (en) * | 2012-04-26 | 2013-10-30 | 上海竞天科技股份有限公司 | Image processing device and image processing method |
US20140307950A1 (en) * | 2013-04-13 | 2014-10-16 | Microsoft Corporation | Image deblurring |
CN103473577B (en) * | 2013-09-06 | 2016-05-11 | 广东华大互联网股份有限公司 | A kind of card sender card Optical Character Recognition system and method |
CN103914816A (en) * | 2014-03-04 | 2014-07-09 | 西安电子科技大学 | Video super-resolution method based on non-local regularization |
GB201410631D0 (en) * | 2014-06-13 | 2014-07-30 | Univ Glasgow | Method and apparatus for evaluating defocus in an image of a scene |
CN108460649A (en) * | 2017-02-22 | 2018-08-28 | 阿里巴巴集团控股有限公司 | A kind of image-recognizing method and device |
-
2019
- 2019-12-23 CN CN201911338461.9A patent/CN111191550B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101996406A (en) * | 2010-11-03 | 2011-03-30 | 中国科学院光电技术研究所 | No-reference structural sharpness image quality evaluation method |
EP2696584A2 (en) * | 2012-08-06 | 2014-02-12 | Ricoh Company, Ltd. | Image capturing device and image capturing method |
CN103279328A (en) * | 2013-04-08 | 2013-09-04 | 河海大学 | BlogRank algorithm parallelization processing construction method based on Haloop |
Also Published As
Publication number | Publication date |
---|---|
CN111191550A (en) | 2020-05-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wang et al. | Adaptive fusion for RGB-D salient object detection | |
CN110210387B (en) | Method, system and device for detecting insulator target based on knowledge graph | |
CN110827312B (en) | Learning method based on cooperative visual attention neural network | |
CN114937232B (en) | Wearing detection method, system and equipment for medical waste treatment personnel protective appliance | |
WO2020029915A1 (en) | Artificial intelligence-based device and method for tongue image splitting in traditional chinese medicine, and storage medium | |
WO2019127102A1 (en) | Information processing method and apparatus, cloud processing device, and computer program product | |
CN106529441B (en) | Depth motion figure Human bodys' response method based on smeared out boundary fragment | |
CN106599806A (en) | Local curved-surface geometric feature-based human body action recognition method | |
WO2013075295A1 (en) | Clothing identification method and system for low-resolution video | |
CN104951440B (en) | Image processing method and electronic equipment | |
CN110826612A (en) | Training and identifying method for deep learning | |
CN111209873A (en) | High-precision face key point positioning method and system based on deep learning | |
CN110826534B (en) | Face key point detection method and system based on local principal component analysis | |
Tripathy | Detection of cotton leaf disease using image processing techniques | |
CN114372962A (en) | Laparoscopic surgery stage identification method and system based on double-particle time convolution | |
CN111191550B (en) | Visual perception device and method based on automatic dynamic adjustment of image sharpness | |
CN113065400A (en) | Invoice seal detection method and device based on anchor-frame-free two-stage network | |
CN112560925A (en) | Complex scene target detection data set construction method and system | |
Zhang et al. | An Improved Computational Approach for Salient Region Detection. | |
CN114449362B (en) | Video cover selection method, device, equipment and storage medium | |
CN113010736A (en) | Video classification method and device, electronic equipment and storage medium | |
CN110598716A (en) | Personnel attribute identification method, device and system | |
CN111476174B (en) | Face image-based emotion recognition method and device | |
CN110276260B (en) | Commodity detection method based on depth camera | |
CN114202723A (en) | Intelligent editing application method, device, equipment and medium through picture recognition |
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 |