CN112465850A - Peripheral boundary modeling method, intelligent monitoring method and device - Google Patents
Peripheral boundary modeling method, intelligent monitoring method and device Download PDFInfo
- Publication number
- CN112465850A CN112465850A CN202011425827.9A CN202011425827A CN112465850A CN 112465850 A CN112465850 A CN 112465850A CN 202011425827 A CN202011425827 A CN 202011425827A CN 112465850 A CN112465850 A CN 112465850A
- Authority
- CN
- China
- Prior art keywords
- current
- boundary
- peripheral boundary
- image
- intelligent monitoring
- 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.)
- Pending
Links
- 230000002093 peripheral effect Effects 0.000 title claims abstract description 45
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000012544 monitoring process Methods 0.000 title claims abstract description 31
- 238000003708 edge detection Methods 0.000 claims abstract description 28
- 238000013507 mapping Methods 0.000 claims abstract description 7
- 238000012806 monitoring device Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 14
- 238000001514 detection method Methods 0.000 description 19
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000010261 cell growth Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention relates to a peripheral boundary modeling method, an intelligent monitoring method and an intelligent monitoring device. The technical scheme of the invention is as follows: the modeling method comprises the following steps: performing edge detection on the image to obtain a binary edge detection result; performing expansion operation on the binary image of the edge detection result; traversing the points in the expanded binary image, and representing the pixel value of the point (x, y) in the image by P (x, y), if the following conditions are satisfied under the same x
Recording the current coordinate (x, y) as a boundary point when the current coordinate (x, y) is 255, ending the current cycle, and continuously judging that the y value corresponding to the next point x +1 isWhether the conditions are met; if the adjacent x position has a boundary point and the current x position does not have a boundary point, acquiring edge information obtained from the previous position by the current x position; and correspondingly writing the recorded data (x, y) into a map mapping table, wherein x is a key value, and y is a corresponding value, and finishing modeling the peripheral boundary. The invention is suitable for the field of intelligent building monitoring.
Description
Technical Field
The invention relates to a peripheral boundary modeling method, an intelligent monitoring method and an intelligent monitoring device. The intelligent monitoring system is suitable for the field of intelligent monitoring of buildings.
Background
With the continuous development of living standard and the rapid progress of society, if the real estate market of China is compared with a water pool, the water level of the water pool continuously and rapidly rises in the last decade, namely, the housing stock of China is greatly increased.
With the continuous growth of cells and high-rise buildings, various hazard problems such as high altitude parabolic and the like continuously occur, the buildings need to be supervised and analyzed for 24 hours, and because most of safety events in the buildings at present occur at high floors, and the safety events are frequently and frequently seen and prohibited. Therefore, the intelligent building is developed to be the current research hotspot, and the computer vision is adopted to analyze and process the building monitoring picture, so that the problem that the manpower cannot be continuously monitored for 24 hours is solved, and the labor cost is reduced.
In the monitoring video, the floor probably only occupies one part of the whole picture, if the whole picture is analyzed, the calculated amount is large, at the moment, the floor part is detected and marked, the calculation pressure can be reduced, meanwhile, the interference of the external environment of the floor on intelligent analysis can be reduced, and the analysis accuracy is improved.
Taking a high-altitude parabolic model as an example, the current monitoring picture needs to be analyzed, at this time, the peripheral detection of the building is particularly important, and after the peripheral range of the building is located, the calculation method analysis can be performed only on the building range, so that the calculation range can be effectively reduced, and the calculation amount is reduced.
Edge detection is to identify the boundary of an image by the variation of a first or second derivative according to the significant change of the attribute in the image, and usually locate the boundary in the direction of the maximum gradient. The edge detection can eliminate information which is considered irrelevant, retain important structural attributes of the image, and play a very important role in a plurality of image processing applications. However, the edges of natural images are not always ideal step edges, but rather they are typically affected by one or more of the following: 1. focus blur due to limited scene depth; 2. penumbra blurring caused by shadows generated by non-zero radius light sources; 3. shadows of smooth object edges; 4. partial specular or diffuse reflection near the edges of the object. Common edge detection operators include Roberts, Prewitt, Sobel, Canny, and the like, and the purpose of the edge detection operators is to find a set formed by pixels with severely changed brightness in an image.
At present, edge detection algorithms are more, but when the edge detection is performed on a building, the existing edge detection algorithms can obtain all edges in an image, the particularity of the surface of the building is not considered, and areas with obvious edges, such as windows and the like, included in a floor are not considered.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the existing problems, a peripheral boundary modeling method, an intelligent monitoring method and an intelligent monitoring device are provided.
The technical scheme adopted by the invention is as follows: a method of peripheral boundary modeling, comprising:
performing edge detection on the image to obtain a binary edge detection result;
performing expansion operation on the binary image of the edge detection result;
traversing the points in the expanded binary image, and representing the pixel value of the point (x, y) in the image by P (x, y), if the following conditions are satisfied under the same x
P(x，y)＝255
Recording the current coordinates (x, y) as boundary points, ending the current cycle, and continuously judging whether the y value corresponding to the next point x +1 meets the conditions; wherein width is the width of the image and height is the height of the image;
if the adjacent x position has a boundary point and the current x position does not have a boundary point, acquiring edge information obtained from the previous position by the current x position;
and correspondingly writing the recorded data (x, y) into a map mapping table, wherein x is a key value, and y is a corresponding value, and finishing modeling the peripheral boundary.
If there is a recorded boundary point in the adjacent x position and there is no boundary point in the current x position, the current x position obtains the edge information obtained from the previous position, including:
if the current x position satisfies the following condition
And the current position does not obtain the corresponding boundary point after traversing, (x)n-1，yn-1) The value of (d) is assigned to the current position.
A storage medium having a computer program stored thereon, characterized in that: the computer program, when executed by a processor, implements the steps of the peripheral boundary modeling method.
An intelligent monitoring method is characterized in that:
acquiring a video frame in a monitoring video;
judging whether the current frame is an initial frame, if so, obtaining peripheral boundary information by the peripheral boundary modeling method of claim 1 or 2, and storing the peripheral boundary information; if not, obtaining the stored peripheral boundary information;
judging whether a target event in the video frame occurs in the peripheral boundary range or not according to the peripheral boundary information, and if so, carrying out subsequent intelligent calculation; if not, follow-up intelligent calculation is not needed.
And analyzing the number of the collected video frame to determine whether the current frame is an initial frame.
A storage medium having a computer program stored thereon, characterized in that: the computer program realizes the steps of the intelligent monitoring method when being executed by a processor.
An intelligent monitoring device having a processor and a memory, the memory having stored thereon a computer program, characterized in that: the computer program realizes the steps of the intelligent monitoring method when being executed by a processor.
And the system also comprises a video shooting module for shooting monitoring videos.
The invention has the beneficial effects that: the invention provides a method for analyzing and calculating whether the obtained boundary information point is a peripheral boundary or not based on the edge detection result, which can completely detect the peripheral boundary and successfully eliminate the influence caused by internal noise.
According to the invention, the detection peripheral boundary modeling is carried out on the initial frame of the monitoring video, and on the premise that the camera does not move, the subsequent frame directly uses the boundary modeling result of the initial frame, so that the repeated calculation is not needed, and the calculation amount is reduced.
The invention judges whether the detection target appears in the peripheral boundary range according to the peripheral boundary information, completes target screening and reduces the calculation amount of subsequent intelligent calculation.
Drawings
Fig. 1 is a flowchart of an intelligent monitoring method in an embodiment.
FIG. 2 is a diagram illustrating edge detection in an embodiment.
FIG. 3 is a diagram illustrating a peripheral boundary detection result in an embodiment.
Detailed Description
The embodiment is a building intelligent monitoring method, which comprises the following steps (see fig. 1):
acquiring a video frame in a building monitoring video;
judging whether the current frame is an initial frame, if so, obtaining and storing peripheral boundary information of the building by a peripheral boundary modeling method; if not, obtaining the stored peripheral boundary information; the present example determines whether the current frame is an initial frame by analyzing the number of the collected video frame, and performs corresponding processing on the current image according to different results;
judging whether a target event in the video frame occurs in the peripheral boundary range or not according to the peripheral boundary information, and if so, carrying out subsequent intelligent calculation; if not, follow-up intelligent calculation is not needed.
The method for modeling the peripheral boundary in the embodiment comprises the following steps:
and performing edge detection on the initial frame in the building monitoring video to obtain a rich and complete image edge detection result. A commonly used edge detection operator can be selected, in this example, a Sobel operator is taken as an example, the operator is a discrete difference operator, convolution kernels in two directions x and y are given, the convolution kernels are used for calculating an approximate value of the gray scale of the image brightness function, and the place where the change of the gray scale is severe is considered to be the edge. Fig. 2 is a diagram of a detection result of the Sobel operator, and for the appearance of a building, a window and the like may affect the detection. Although the operator can obtain complete and detailed edge information, when the peripheral boundary of the building is located, the edge detection result brought by a window and the like belongs to useless detection information.
After obtaining the binary edge detection result, in order to make the finally obtained boundary continuous and complete, the obtained binary image is subjected to an expansion operation, that is, a white part (edge part) in the image is expanded, the expanded image has a highlight area larger than that of the original image, noise is eliminated, and discontinuous parts in the detection result binary image are connected.
After obtaining the result of the expanded edge detection binary image, traversing points in the image, and representing the pixel value of the point (x, y) in the image by P (x, y), wherein in the binary image of the edge detection result, the pixel values are only 0 or 255, wherein 0 represents black, 255 represents white, the pixel point of the edge point is 255, wherein x represents abscissa, y represents ordinate, traversing is started from the upper left corner or the upper right corner of the image, and if the condition that the pixel values satisfy the condition of the same x is met
P(x，y)＝255
Then recording the current coordinate (x, y), ending the current cycle, and meanwhile, continuously judging whether the y value corresponding to the next point x +1 meets the above condition. When a plurality of y correspond to each other, selecting the minimum y value, namely selecting the boundary point of the outermost periphery in the image as the boundary point corresponding to the current x; meanwhile, the algorithm does not consider the case that the peripheral boundary is vertical, because when the boundary is a vertical straight line, no detected area exists, and the determination method of the boundary has no meaning.
In order to prevent the discontinuity problem in the edge detection result image, whether edge points exist in adjacent x positions or not is analyzed, if the adjacent x positions exist and the edge points do not exist in the current x position, edge information obtained from the previous position is obtained, and therefore the integrity of the image edge detection result is guaranteed. That is, if the current x position is determined to satisfy the following condition
And the current position does not obtain the corresponding boundary point after traversing, and (x)n-1，yn-1) The value of (2) is assigned to the current position to ensure continuity of the image boundary. Corresponding the calculated data (x, y)And writing the data into a map mapping table, wherein x is a key value, and y is a corresponding value, namely completing building boundary modeling. The map mapping table is traversed to obtain the current building peripheral boundary information map, as shown in fig. 3 below.
The method for judging whether the target event occurs in the video frame within the peripheral boundary range in the embodiment includes:
after the map mapping table of the peripheral boundary information is obtained, when the intelligent building detection is carried out, whether the current event occurs in the building range can be judged through the mapping table, taking a high-altitude parabolic event as an example, the detection algorithm marks the result obtained by monitoring by using a rectangular frame, that is, the position of the object is marked by a rectangular frame in the image, if the information of the position of the object is (x, y, width, height), wherein x and y are coordinate information of the upper left corner of the parabolic detection frame, width is width of the rectangular frame of the parabolic detection result, and height is height of the rectangular frame of the parabolic detection result (the detection frame here refers to target position information detected by an algorithm, taking parabolic detection as an example, when a parabolic event occurs, the algorithm detects and marks a parabolic object with the rectangular frame in a picture, and the rectangular frame is the detection frame), and whether the obtained detection result is in the building range is judged by the following formula.
Wherein valuexRepresenting the corresponding building peripheral boundary point, value, taken on the abscissa x(x+width)And if the condition is not met, the obtained position information is not in the building range, and the position information can be deleted, so that the calculation amount of subsequent intelligent calculation (intelligent calculation for parabolic detection and the like) is reduced, the interference caused by the environment is reduced, and the judgment and analysis accuracy is improved.
The embodiment also provides an intelligent monitoring device, which comprises a video shooting module (monitoring camera), a processor and a memory, wherein the memory is stored with a computer program, the video shooting module is used for shooting the monitoring video of the monitored building, and the computer program is executed by the processor to realize the steps of the intelligent monitoring method.
The method comprises the steps of firstly judging an effective building boundary by analyzing the edge detection result of a video starting frame, positioning an effective calculation range, and analyzing whether an intelligent detection result in a current frame is within the building range or not by a subsequent frame according to the obtained boundary positioning result. According to the method, the effective calculation range is located, so that the intelligent calculation amount is reduced, meanwhile, the influence of the environment outside the building on the intelligent calculation result can be effectively removed by locating the boundary, and the calculation accuracy is improved.
Claims (8)
1. A method of peripheral boundary modeling, comprising:
performing edge detection on the image to obtain a binary edge detection result;
performing expansion operation on the binary image of the edge detection result;
traversing the points in the expanded binary image, and representing the pixel value of the point (x, y) in the image by P (x, y), if the following conditions are satisfied under the same x
P(x，y)＝255
Recording the current coordinates (x, y) as boundary points, ending the current cycle, and continuously judging whether the y value corresponding to the next point x +1 meets the conditions; wherein width is the width of the image and height is the height of the image;
if the adjacent x position has a boundary point and the current x position does not have a boundary point, acquiring edge information obtained from the previous position by the current x position;
and writing the recorded data (x, y) into a map mapping table in a one-to-one mode, wherein x is a key value, and y is a corresponding value, and finishing modeling the peripheral boundary.
2. The peripheral boundary modeling method of claim 1, wherein if there are boundary points recorded in adjacent x positions and there are no boundary points in the current x position, the current x position obtains the edge information from the previous position, including:
if the current x position satisfies the following condition
And the current position does not obtain the corresponding boundary point after traversing, (x)n-1，yn-1) The value of (d) is assigned to the current position.
3. A storage medium having a computer program stored thereon, characterized in that: the computer program, when being executed by a processor, realizes the steps of the peripheral boundary modeling method of claim 1 or 2.
4. An intelligent monitoring method is characterized in that:
acquiring a video frame in a monitoring video;
judging whether the current frame is an initial frame, if so, obtaining peripheral boundary information by the peripheral boundary modeling method of claim 1 or 2, and storing the peripheral boundary information; if not, obtaining the stored peripheral boundary information;
judging whether a target event in the video frame occurs in the peripheral boundary range or not according to the peripheral boundary information, and if so, carrying out subsequent intelligent calculation; if not, follow-up intelligent calculation is not needed.
5. The intelligent monitoring method according to claim 4, wherein: and analyzing the number of the collected video frame to determine whether the current frame is an initial frame.
6. A storage medium having a computer program stored thereon, characterized in that: the computer program, when being executed by a processor, realizes the steps of the intelligent monitoring method of claim 4 or 5.
7. An intelligent monitoring device having a processor and a memory, the memory having stored thereon a computer program, characterized in that: the computer program, when being executed by a processor, realizes the steps of the intelligent monitoring method of claim 4 or 5.
8. The intelligent monitoring device of claim 7, wherein: and the system also comprises a video shooting module for shooting monitoring videos.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011425827.9A CN112465850A (en) | 2020-12-08 | 2020-12-08 | Peripheral boundary modeling method, intelligent monitoring method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011425827.9A CN112465850A (en) | 2020-12-08 | 2020-12-08 | Peripheral boundary modeling method, intelligent monitoring method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112465850A true CN112465850A (en) | 2021-03-09 |
Family
ID=74800295
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011425827.9A Pending CN112465850A (en) | 2020-12-08 | 2020-12-08 | Peripheral boundary modeling method, intelligent monitoring method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112465850A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113298046A (en) * | 2021-06-29 | 2021-08-24 | 中国科学院计算技术研究所数字经济产业研究院 | High-altitude parabolic detection method based on monitoring video |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103077540A (en) * | 2013-01-30 | 2013-05-01 | 中国科学院自动化研究所 | Automatic detection method of vehicle moving region in video monitoring of express way |
| CN110414333A (en) * | 2019-06-20 | 2019-11-05 | 平安科技（深圳）有限公司 | A kind of detection method and device of image boundary |
| CN111260675A (en) * | 2020-01-21 | 2020-06-09 | 武汉大学 | High-precision extraction method and system for image real boundary |
| CN111681256A (en) * | 2020-05-07 | 2020-09-18 | 浙江大华技术股份有限公司 | Image edge detection method and device, computer equipment and readable storage medium |
-
2020
- 2020-12-08 CN CN202011425827.9A patent/CN112465850A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103077540A (en) * | 2013-01-30 | 2013-05-01 | 中国科学院自动化研究所 | Automatic detection method of vehicle moving region in video monitoring of express way |
| CN110414333A (en) * | 2019-06-20 | 2019-11-05 | 平安科技（深圳）有限公司 | A kind of detection method and device of image boundary |
| CN111260675A (en) * | 2020-01-21 | 2020-06-09 | 武汉大学 | High-precision extraction method and system for image real boundary |
| CN111681256A (en) * | 2020-05-07 | 2020-09-18 | 浙江大华技术股份有限公司 | Image edge detection method and device, computer equipment and readable storage medium |
Non-Patent Citations (3)
| Title |
|---|
| 冯德俊等: "动态监测变化图斑边界跟踪", 《遥感技术与应用》 * |
| 周亚罗等: "基于图像处理的轮廓提取方法应用", 《唐山学院学报》 * |
| 张世辉等: "基于无监督在线学习实现视频遮挡边界检测", 《光学学报》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113298046A (en) * | 2021-06-29 | 2021-08-24 | 中国科学院计算技术研究所数字经济产业研究院 | High-altitude parabolic detection method based on monitoring video |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113781402B (en) | Method and device for detecting scratch defects on chip surface and computer equipment | |
| EP2375376A1 (en) | Method and arrangement for multi-camera calibration | |
| CN110473221B (en) | Automatic target object scanning system and method | |
| CN110766720A (en) | Multi-camera vehicle tracking system based on deep learning | |
| CN110610150A (en) | Tracking method, device, computing equipment and medium of target moving object | |
| CN111079518A (en) | Fall-down abnormal behavior identification method based on scene of law enforcement and case handling area | |
| CN117252868B (en) | Direct current screen defect detection method based on machine vision | |
| CN111368785A (en) | Camera shielding judgment method, device, equipment and storage medium | |
| JP2006039689A (en) | Image processor, image processing method, image processing program, and recording medium with the program recorded thereon | |
| CN114359412A (en) | Automatic calibration method and system for external parameters of camera facing to building digital twins | |
| CN110675425A (en) | Video frame identification method, device, equipment and medium | |
| CN116777877A (en) | Circuit board defect detection method, device, computer equipment and storage medium | |
| WO2022091577A1 (en) | Information processing device and information processing method | |
| CN114674826A (en) | Visual detection method and detection system based on cloth | |
| CN112465850A (en) | Peripheral boundary modeling method, intelligent monitoring method and device | |
| CN117671299A (en) | Loop detection method, device, equipment and storage medium | |
| CN106778822B (en) | Image straight line detection method based on funnel transformation | |
| CN111667509A (en) | Method and system for automatically tracking moving target under condition that target is similar to background color | |
| CN116993654B (en) | Camera module defect detection method, device, equipment, storage medium and product | |
| CN116052120A (en) | Excavator night object detection method based on image enhancement and multi-sensor fusion | |
| CN112598738B (en) | Character positioning method based on deep learning | |
| CN115205793A (en) | Electric power machine room smoke detection method and device based on deep learning secondary confirmation | |
| CN112132767A (en) | Solar cell panel shadow processing method based on computer vision | |
| CN118134910B (en) | Defect detection method and system for production of liquid leakage sensing assembly | |
| CN116538953B (en) | Intelligent detection method and system for elevation targets and readable storage medium |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210309 |