CN113411571B - Video frame definition detection method based on sliding window gradient entropy - Google Patents
Video frame definition detection method based on sliding window gradient entropy Download PDFInfo
- Publication number
- CN113411571B CN113411571B CN202110665603.3A CN202110665603A CN113411571B CN 113411571 B CN113411571 B CN 113411571B CN 202110665603 A CN202110665603 A CN 202110665603A CN 113411571 B CN113411571 B CN 113411571B
- Authority
- CN
- China
- Prior art keywords
- video frame
- definition
- frame
- gradient
- period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a video frame definition detection method based on sliding window gradient entropy, aiming at a video stream continuously input by a monitoring camera, triggering a video frame definition anomaly detection algorithm at the starting moment of each period of time; extracting a reference video frame and a video frame to be detected; image blocking is carried out on the extracted video frames, and a gradient entropy matrix is constructed; calculating a gradient entropy matrix reference value of the definition normal video frame of each period; calculating the abnormal condition of the definition of the image block of the video frame to be detected and outputting a definition abnormal measurement matrix; and judging the blurring condition of the video frame according to the definition anomaly metric matrix. The method can judge the local definition abnormal condition of the video frame to be detected and mark the abnormal position of the video frame, and can accurately and rapidly detect the video frame with abnormal definition.
Description
Technical Field
The invention relates to the field of image processing, in particular to a video frame definition detection method based on sliding window gradient entropy.
Background
Video surveillance systems are an important part of security technology protection systems, but due to the presence of a large number of blurred images, video availability is greatly reduced, and even many blurred images can bring incorrect information to people. Therefore, it is important to perform real-time image sharpness detection on a video monitoring system.
Currently, there are two methods for quality evaluation of images. Subjective and objective evaluations, respectively. Subjective assessment is the direct judgment of images from human visual effects. The objective evaluation is to judge the image through the establishment of a mathematical model. The current common image objective evaluation methods are 3, namely: full reference quality assessment, weak reference quality assessment, no reference quality assessment.
In the reference-free quality evaluation, based on the traditional digital image definition evaluation method, the definition algorithm which is common and has representativeness at present is an evaluation method based on a gradient function, a Brenne gradient function method, a Tenengrad, laplacian gradient function method and an energy gradient function method.
While these algorithms all have good performance, they all have a common problem. Because these algorithms are all based on the calculation or comparison of the gradients of the whole image, if the image to be judged has local abnormal conditions such as full black or full white of local area pixels, local blurring, local shielding, local brightness abnormality and the like, the traditional algorithm can detect the clear images as abnormal images, so that detection errors are caused. For example, if a part of an image is black and blocked, but the definition of the other part is normal, if the definition of the image is abnormal by obtaining the gradient of the whole image, the gradient of the blocked part affects the gradient of the whole image, so that the obtained gradient value cannot reflect the gradient value of the whole image, and a result misjudgment is formed.
Disclosure of Invention
The invention aims to provide a video frame definition detection method based on sliding window gradient entropy, which can judge the local definition abnormal condition of a video frame to be detected, mark the abnormal position of the video frame to be detected and accurately and rapidly detect the video frame with definition abnormality.
The technical scheme adopted by the invention is as follows:
a video frame definition detection method based on sliding window gradient entropy comprises the following steps:
s1: triggering video frame definition anomaly detection at the starting moment of each set period of time aiming at a video stream continuously input by a monitoring camera;
s2: extracting a reference video frame and a video frame to be detected: continuously extracting a plurality of video frames at the starting moment of each period as reference video frames for calculating definition measurement characteristic reference values; extracting a plurality of video frames in the video stream as video frames to be detected according to a set extraction period in each period;
s3: image blocking is carried out on the extracted video frames, and a gradient entropy matrix is constructed;
s4: calculating and obtaining the average result of the gradient entropy matrix of the reference video frame;
s5: judging whether a reference value exists in the current time period or not; if yes, executing step S6; otherwise, taking the average result of the gradient entropy matrix of the reference video frame as the current reference value and executing the step S7;
s6: comparing each element of the average gradient entropy matrix of the reference video frame with the reference value of the current corresponding period respectively to judge whether the average gradient entropy matrix is in the allowable variation range; if yes, weighting and combining the current reference value with the original element value; otherwise, the original element value is kept as the current reference value;
s7: calculating the abnormal condition of the definition of the image block of the video frame to be detected and outputting a definition abnormal measurement matrix: comparing the gradient entropy matrix of the video frame to be detected in each period with the reference value in the period in element level, and judging the video frame to be detected as a normal image block when the element value of the gradient entropy matrix is in the range of the allowable variation range of the reference value; otherwise, judging the image block to be an abnormal image block, and constructing and outputting a definition abnormal measurement matrix according to whether the image block is abnormal or not and setting a zone bit.
Further, as a preferred embodiment, S1 is a period of one time per hour.
Further, as a preferred embodiment, S2 specifically includes the following steps:
s2-1: in the t-th period, K frames are extracted as reference frames to form a reference frame set F K ={f 1 ,…,f K Of f, where f 1 ,…,f K The first K frames of video frames representing the starting time of the period;
s2-2: in the t-th period, the frame interval to be detected is set to be deltaf, i.e. one frame is extracted as the frame to be detected every time deltaf, and the frame to be detected is set to be F D ={f K+1+Δf ,…,f K+1+n·Δf And n represents the extracted nth frame to be detected.
Further, as a preferred embodiment, in S3, converting the video frame into a gray scale image, and sliding and clipping from left to right and from top to bottom according to a set sliding window to obtain a plurality of image blocks; and calculating gradient entropy of each image block, and constructing a frame gradient entropy matrix according to the position of the image block in the corresponding video frame gray level image.
Further, as a preferred embodiment, S3 specifically includes:
s3-1: for any frame f in the reference frame set k (k=1, …, K) or any frame f in the set of frames to be detected d (d=k+1+Δf, …, k+1+n·Δf) which is converted into a corresponding gray-scale image g k G d ;
S3-2: setting the size of a sliding window as h x w pixels, the transverse sliding step length as x and the longitudinal sliding y; then the gray image g is displayed from left to right and from top to bottom k G d Sliding cutting to obtain M image blocks, each image block being marked as G k,i G (G) d,i Where i=1, 2, …, M; the transverse sliding step length is x, and the longitudinal sliding y is calculated as follows:
wherein, alpha and beta are respective step control coefficients;
s3-3: will G k,i G (G) d Collectively denoted as G i The Sobel kernels in the horizontal and vertical directions are denoted S, respectively x And S is y Calculating the gradient d in the horizontal and vertical directions x And d y G is then i Gradient d of (2) i Is that
S3-4: image block G i Gradient d of (2) i There are N gradient values in total, and the proportion of each gradient value is ρ n (n=1, 2, …, N), d is calculated according to the following formula i Gradient entropy xi of (2) i :
Thereby obtaining G k,i G (G) d,i Gradient entropy xi of (2) k,i ,ξ d,i ;
S3-5: g is determined according to the position of the sliding window k,i G (G) d,i Gradient entropy xi of (2) k,i And xi d,i Constructed as video frame f k And f d Is respectively recorded as the gradient entropy matrix of the Xi k ,Ξ d
Further, as a preferred embodiment, the period K frame is calculated in S4 as the Σ of the reference frame k Mean value xi K :
Further, as a preferred embodiment, S6 specifically includes the following steps:
s6-1: recording the gradient entropy matrix reference value of the definition normal video frame in the t period as the Xi c
S6-2: judging whether a gradient entropy matrix reference value of the definition normal video frame in the t period is assigned; xi when t-th period c Non-assigned xi c =Ξ K The method comprises the steps of carrying out a first treatment on the surface of the Otherwise, for xi c Each element ζ in (a) c,i I=1, 2, …, M, updated according to the following formula:
wherein, gamma c For weighting parameters, sigma c Fault tolerant coefficients are for the gradient entropy reference.
Further, as a preferred embodiment, the specific step S7 includes:
s7-1: defining a video frame f to be detected d Sharpness anomaly metric matrix is ψ d
Wherein the element psi d,i Representing video frame f d The i-th image block definition abnormal flag bit; when the image of the image block is clear, the element value is 0; otherwise, 1;
s7-2: to-be-detected video frame f d Gradient entropy matrix xi d With xi c Element level comparison is performed and ψ is calculated according to the following formula d,i Values of (2)
Wherein sigma d And the abnormal fault tolerance coefficient of the image block definition.
By adopting the technical scheme, the method and the device can judge the local definition abnormal condition of the video frame to be detected, mark the abnormal position of the video frame to be detected and accurately and rapidly detect the video frame with the definition abnormal condition.
Drawings
The invention is described in further detail below with reference to the drawings and detailed description;
FIG. 1 is a flow chart of a method for detecting video frame definition based on sliding window gradient entropy;
FIG. 2 is a schematic diagram of the gradient entropy matrix reference value calculation according to the present invention;
FIG. 3 is a schematic diagram of a video frame sharpness anomaly metrics matrix according to the present invention.
Detailed Description
For the purposes, technical solutions and advantages of the embodiments of the present application, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.
As shown in one of fig. 1 to 3, the invention discloses a method for detecting video frame definition based on sliding window gradient entropy, which comprises the following steps:
s1: triggering video frame definition anomaly detection at the starting moment of each set period of time aiming at a video stream continuously input by a monitoring camera;
further, as a preferred embodiment, S1 is a period of one time per hour. And creating a folder to record the total video file after the input video stream is divided into frames for 5s, and further, regularly naming pictures after the video is stored into a designated folder in a frame dividing manner.
S2: extracting a reference video frame and a video frame to be detected: continuously extracting a plurality of video frames at the starting moment of each period as reference video frames for calculating definition measurement characteristic reference values; extracting a plurality of video frames in the video stream as video frames to be detected according to a set extraction period in each period; s2 specifically comprises the following steps:
s2-1: in the t-th period, K frames are extracted as reference frames to form a reference frame set F K ={f 1 ,…,f K Of f, where f 1 ,…,f K The first K frames of video frames representing the starting time of the period;
s2-2: in the t-th period, the frame interval to be detected is set to be deltaf, i.e. one frame is extracted as the frame to be detected every time deltaf, and the frame to be detected is set to be F D ={f K+1+Δf ,…,f K+1+n·Δf And n represents the extracted nth frame to be detected.
S3: image blocking is carried out on the extracted video frames, and a gradient entropy matrix is constructed;
specifically, converting a video frame into a gray level image, and sliding and cutting from left to right and from top to bottom according to a set sliding window to obtain a plurality of image blocks; calculating gradient entropy of each image block, and constructing a frame gradient entropy matrix according to the position of the image block in the corresponding video frame gray level image; s3 specifically comprises the following steps:
s3-1: for any frame f in the reference frame set k (k=1, …, K) or any frame f in the set of frames to be detected d (d=k+1+Δf, …, k+1+n·Δf) which is converted into a corresponding gray-scale image g k G d ;
S3-2: setting the size of a sliding window as h x w pixels, the transverse sliding step length as x and the longitudinal sliding y; then the gray image g is displayed from left to right and from top to bottom k G d Sliding cutting to obtain M image blocks, each image block being marked as G k,i G (G) d,i Where i=1, 2, …, M; the transverse sliding step length is x, and the longitudinal sliding y is calculated as follows:
wherein, alpha and beta are respective step control coefficients;
s3-3: will G k,i G (G) d Collectively denoted as G i The Sobel kernels in the horizontal and vertical directions are denoted S, respectively x And S is y Calculating the gradient d in the horizontal and vertical directions x And d y G is then i Gradient d of (2) i Is that
S3-4: image block G i Gradient d of (2) i There are N gradient values in total, and the proportion of each gradient value is ρ n (n=1, 2, …, N), d is calculated according to the following formula i Gradient entropy xi of (2) i :
Thereby obtaining G k,i G (G) d,i Gradient entropy xi of (2) k,i ,ξ d,i ;
S3-5: g is determined according to the position of the sliding window k,i G (G) d,i Gradient entropy xi of (2) k,i And xi d,i Constructed as video frame f k And f d Is respectively recorded as the gradient entropy matrix of the Xi k ,Ξ d
S4: calculating and obtaining the average result of the gradient entropy matrix of the reference video frame; further, as a preferred embodiment, the period K frame is calculated in S4 as the Σ of the reference frame k Mean value of
S5: as shown in fig. 2, judging whether a reference value exists in the current period; if yes, executing step S6; otherwise, taking the average result of the gradient entropy matrix of the reference video frame as the current reference value and executing the step S7;
s6: comparing each element of the average gradient entropy matrix of the reference video frame with the reference value of the current corresponding period respectively to judge whether the average gradient entropy matrix is in the allowable variation range; if yes, weighting and combining the current reference value with the original element value; otherwise, the original element value is kept as the current reference value;
specifically, S6 includes the steps of:
s6-1: recording the gradient entropy matrix reference value of the definition normal video frame in the t period as the Xi c
S6-2: judging gradient entropy matrix of normal video frame with definition in t periodWhether the reference value is assigned; xi when t-th period c Is not assigned, thenOtherwise, for xi c Each element ζ in (a) c,i I=1, 2, …, M, updated according to the following formula:
wherein, gamma c For weighting parameters, sigma c Fault tolerant coefficients are for the gradient entropy reference.
S7: calculating the abnormal condition of the definition of the image block of the video frame to be detected and outputting a definition abnormal measurement matrix: comparing the gradient entropy matrix of the video frame to be detected in each period with the reference value in the period in element level, and judging the video frame to be detected as a normal image block when the element value of the gradient entropy matrix is in the range of the allowable variation range of the reference value; otherwise, judging the image block to be an abnormal image block, and constructing and outputting a definition abnormal measurement matrix according to whether the image block is abnormal or not and setting a zone bit.
Further, as shown in fig. 3, as a preferred embodiment, the specific steps of S7 include:
s7-1: defining a video frame f to be detected d Sharpness anomaly metric matrix is ψ d
Wherein the element psi d,i Representing video frame f d The i-th image block definition abnormal flag bit; when the image of the image block is clear, the element value is 0; otherwise, 1;
s7-2: to-be-detected video frame f d Gradient entropy matrix xi d With xi c Element level comparison is performed and ψ is calculated according to the following formula d,i Values of (2)
Wherein sigma d And the abnormal fault tolerance coefficient of the image block definition.
S8: and judging the blurring condition of the video frame according to the definition anomaly metric matrix. If the image is a local abnormal image, outputting a local abnormal position; if the image is a global blurred image, the output is a blurred image.
By adopting the technical scheme, the method and the device can judge the local definition abnormal condition of the video frame to be detected, mark the abnormal position of the video frame to be detected and accurately and rapidly detect the video frame with the definition abnormal condition.
It will be apparent that the embodiments described are some, but not all, of the embodiments of the present application. Embodiments and features of embodiments in this application may be combined with each other without conflict. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the detailed description of the embodiments of the present application is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
Claims (7)
1. A video frame definition detection method based on sliding window gradient entropy is characterized by comprising the following steps: which comprises the following steps:
s1: triggering video frame definition anomaly detection at the starting moment of each set period of time aiming at a video stream continuously input by a monitoring camera;
s2: extracting a reference video frame and a video frame to be detected: continuously extracting a plurality of video frames at the starting moment of each period as reference video frames for calculating definition measurement characteristic reference values; extracting a plurality of video frames in the video stream as video frames to be detected according to a set extraction period in each period; s2 specifically comprises:
s2-1: at the t-th period, let the extracted K frame be used as the referenceFrames, constituting a reference frame set F K ={f 1 ,…,f K Of f, where f 1 ,…,f K The first K frames of video frames representing the starting time of the period;
s2-2: in the t-th period, the frame interval to be detected is set to be deltaf, i.e. one frame is extracted as the frame to be detected every time deltaf, and the frame to be detected is set to be F D ={f K+1+Δf ,…,f K+1+n·Δf -wherein n represents the extracted nth frame to be detected;
s3: image blocking is carried out on the extracted video frames, and a gradient entropy matrix is constructed; s3 specifically comprises the following steps:
s3-1: for any frame f in the reference frame set k K=1, …, K; or any frame f in the frame set to be detected d D=k+1+Δf, …, k+1+n·Δf; converts it into a corresponding gray-scale image g k G d ;
S3-2: setting the size of a sliding window as h x w pixels, the transverse sliding step length as x and the longitudinal sliding y; then the gray image g is displayed from left to right and from top to bottom k G d Sliding cutting to obtain M image blocks, each image block being marked as G k,i G (G) d,i Where i=1, 2, …, M; the transverse sliding step length is x, and the longitudinal sliding y is calculated as follows:
wherein, alpha and beta are respective step control coefficients;
s3-3: will G k,i G (G) d Collectively denoted as G i The Sobel kernels in the horizontal and vertical directions are denoted S, respectively x And S is y Calculating the gradient d in the horizontal and vertical directions x And d y G is then i Gradient d of (2) i The method comprises the following steps:
s3-4: image block G i Gradient d of (2) i There are N gradient values in total, and the proportion of each gradient value is ρ n (n=1, 2, …, N), d is calculated according to the following formula i Gradient entropy xi of (2) i :
Thereby obtaining G k,i G (G) d,i Gradient entropy xi of (2) k,i ,ξ d,i ;
S3-5: g is determined according to the position of the sliding window k,i G (G) d,i Gradient entropy xi of (2) k,i And xi d,i Constructed as video frame f k And f d Is respectively recorded as the gradient entropy matrix of the Xi k ,Ξ d
S4: calculating and obtaining the average result of the gradient entropy matrix of the reference video frame;
s5: judging whether a reference value exists in the current time period or not; if yes, executing step S6; otherwise, taking the average result of the gradient entropy matrix of the reference video frame as the current reference value and executing the step S7;
s6: comparing each element of the average gradient entropy matrix of the reference video frame with the reference value of the current corresponding period respectively to judge whether the average gradient entropy matrix is in the allowable variation range; if yes, weighting and combining the current reference value with the original element value; otherwise, the original element value is kept as the current reference value;
s7: calculating the abnormal condition of the definition of the image block of the video frame to be detected and outputting a definition abnormal measurement matrix: comparing the gradient entropy matrix of the video frame to be detected in each period with the reference value in the period in element level, and judging the video frame to be detected as a normal image block when the element value of the gradient entropy matrix is in the range of the allowable variation range of the reference value; otherwise, judging the image block to be an abnormal image block, and constructing and outputting a definition abnormal measurement matrix according to whether the image block is abnormal or not and setting a zone bit.
2. The method for detecting video frame definition based on sliding window gradient entropy according to claim 1, wherein the method comprises the following steps: in S1, each hour is a period of time.
3. The method for detecting video frame definition based on sliding window gradient entropy according to claim 1, wherein the method comprises the following steps: s3, converting the video frame into a gray level image, and sliding and cutting from left to right and from top to bottom according to a set sliding window to obtain a plurality of image blocks; and calculating gradient entropy of each image block, and constructing a frame gradient entropy matrix according to the position of the image block in the corresponding video frame gray level image.
5. The method for detecting video frame definition based on sliding window gradient entropy according to claim 1, wherein the method comprises the following steps: s6 specifically comprises the following steps:
s6-1: recording the gradient entropy matrix reference value of the definition normal video frame in the t period as the Xi c
S6-2: judging that the gradient entropy matrix reference value of the definition normal video frame in the t period isWhether to be assigned; xi when t-th period c Is not assigned, thenOtherwise, for xi c Each element ζ in (a) c,i I=1, 2, …, M, updated according to the following formula:
wherein, gamma c For weighting parameters, sigma c Fault tolerant coefficients are for the gradient entropy reference.
6. The method for detecting video frame definition based on sliding window gradient entropy according to claim 1, wherein the method comprises the following steps: s7 specifically comprises the following steps:
s7-1: defining a video frame f to be detected d Sharpness anomaly metric matrix is ψ d
Wherein the element psi d,i Representing video frame f d The i-th image block definition abnormal flag bit; when the image of the image block is clear, the element value is 0; otherwise, 1;
s7-2: to-be-detected video frame f d Gradient entropy matrix xi d With xi c Element level comparison is performed and ψ is calculated according to the following formula d,i Values of (2)
Wherein sigma d And the abnormal fault tolerance coefficient of the image block definition.
7. The method for detecting video frame definition based on sliding window gradient entropy according to claim 1, wherein the method comprises the following steps: it also comprises the following steps:
s8: judging the blurring condition of the video frame according to the definition anomaly metric matrix: outputting a local abnormal position when the local abnormal image is judged; when the global blurred image is judged, the global blurred image is output as the blurred image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110665603.3A CN113411571B (en) | 2021-06-16 | 2021-06-16 | Video frame definition detection method based on sliding window gradient entropy |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110665603.3A CN113411571B (en) | 2021-06-16 | 2021-06-16 | Video frame definition detection method based on sliding window gradient entropy |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113411571A CN113411571A (en) | 2021-09-17 |
CN113411571B true CN113411571B (en) | 2023-07-14 |
Family
ID=77684449
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110665603.3A Active CN113411571B (en) | 2021-06-16 | 2021-06-16 | Video frame definition detection method based on sliding window gradient entropy |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113411571B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115687677B (en) * | 2022-10-18 | 2024-03-22 | 浙江丰能医药科技有限公司 | Data storage management method and system based on artificial intelligence |
CN115866411B (en) * | 2023-03-01 | 2023-05-12 | 深圳市昂星科技有限公司 | Vehicle-mounted monitoring self-adaptive exposure method, device and equipment based on light correction |
CN116866665B (en) * | 2023-09-05 | 2023-11-14 | 中信建投证券股份有限公司 | Video playing method and device, electronic equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2008106994A1 (en) * | 2007-03-07 | 2008-09-12 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Concept for determining a video quality measure for block coded images |
CN104301722A (en) * | 2014-11-03 | 2015-01-21 | 电子科技大学 | Video streaming fuzzy detection method based on frequency domain |
US9232118B1 (en) * | 2015-01-23 | 2016-01-05 | Interra Systems, Inc | Methods and systems for detecting video artifacts |
JP6578619B1 (en) * | 2019-04-11 | 2019-09-25 | 株式会社ラムテック | Image analysis apparatus and image analysis program |
CN112818737A (en) * | 2020-12-18 | 2021-05-18 | 广州视源电子科技股份有限公司 | Video identification method and device, storage medium and terminal |
-
2021
- 2021-06-16 CN CN202110665603.3A patent/CN113411571B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2008106994A1 (en) * | 2007-03-07 | 2008-09-12 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Concept for determining a video quality measure for block coded images |
CN104301722A (en) * | 2014-11-03 | 2015-01-21 | 电子科技大学 | Video streaming fuzzy detection method based on frequency domain |
US9232118B1 (en) * | 2015-01-23 | 2016-01-05 | Interra Systems, Inc | Methods and systems for detecting video artifacts |
JP6578619B1 (en) * | 2019-04-11 | 2019-09-25 | 株式会社ラムテック | Image analysis apparatus and image analysis program |
CN112818737A (en) * | 2020-12-18 | 2021-05-18 | 广州视源电子科技股份有限公司 | Video identification method and device, storage medium and terminal |
Non-Patent Citations (2)
Title |
---|
SHAPE GRADIENT FOR IMAGE SEGMENTATION USING INFORMATION THEORY;A.Herbulot 等;《 ICASSP 2004》;全文 * |
模糊图像评价及在视频去模糊中的应用;李鹏程;《硕士学位论文》;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN113411571A (en) | 2021-09-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113411571B (en) | Video frame definition detection method based on sliding window gradient entropy | |
US11800969B2 (en) | Method and device for monitoring colonoscope withdrawal speed | |
JP4663756B2 (en) | Abnormal behavior detection device | |
JP5701005B2 (en) | Object detection apparatus, object detection method, surveillance camera system, and program | |
US11657491B2 (en) | Learning data collection apparatus, learning data collection method, and program | |
US8630453B2 (en) | Image processing device, image processing method and program | |
US10853949B2 (en) | Image processing device | |
JP6565600B2 (en) | Attention detection device and attention detection method | |
US20180352177A1 (en) | Image-processing device | |
US20150213621A1 (en) | Fire detection system and method employing digital images processing | |
CN112367520B (en) | Video quality diagnosis system based on artificial intelligence | |
CN115862259B (en) | Fire alarm early warning system based on temperature monitoring | |
CN109117774B (en) | Multi-view video anomaly detection method based on sparse coding | |
KR101750094B1 (en) | Method for classification of group behavior by real-time video monitoring | |
Lin et al. | Cross camera people counting with perspective estimation and occlusion handling | |
CN111967345B (en) | Method for judging shielding state of camera in real time | |
RU2419150C1 (en) | Method to process sequence of images to detect and follow air objects | |
JP6214426B2 (en) | Object detection device | |
US20200013172A1 (en) | Object tracking device and object tracking method | |
KR101357721B1 (en) | Apparatus for detecting of single elderly persons’ abnormal situation using image and method thereof | |
US10728476B2 (en) | Image processing device, image processing method, and image processing program for determining a defective pixel | |
CN114205578A (en) | Video fuzzy anomaly detection method and system based on frequency domain skewness and frequency domain peak value | |
JP6904062B2 (en) | Information processing equipment, information processing methods and programs | |
US10277926B2 (en) | Block-base error measure method for object segmentation | |
JP5599228B2 (en) | Busy detection system and busy detection program |
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 |