CN105654503A - Dynamic target detection method based on video images - Google Patents
Dynamic target detection method based on video images Download PDFInfo
- Publication number
- CN105654503A CN105654503A CN201410632496.4A CN201410632496A CN105654503A CN 105654503 A CN105654503 A CN 105654503A CN 201410632496 A CN201410632496 A CN 201410632496A CN 105654503 A CN105654503 A CN 105654503A
- Authority
- CN
- China
- Prior art keywords
- target
- images
- region
- area
- moving objects
- 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
Landscapes
- Image Analysis (AREA)
- Picture Signal Circuits (AREA)
Abstract
The invention discloses a dynamic target detection method based on video images. The method comprises the following steps: performing filtering on input images to remove noise; by use of three adjacent frames of images, respectively performing differentiating on an intermediate frame with adjacent frames, and then performing AND operation on results; and removing noise points and false target areas in binary images, and then improving the boundary of a real target area by use of an area growth method. According to the invention, differentiating is performed on the adjacent image frames firstly, binarization is performed on differential gray-scale images, then, the binary images are scanned for the purpose of removing false moving objects and isolated bright spots, and finally, information of moving objects is improved by use of the area growth method. Each moving target can be accurately detected in a video with a complex background and multiple moving objects, cavities of the targets are effectively filled, and compared to a conventional frame differential method, a conventional teaching morphological method and the like, the moving objects can be more accurately detected.
Description
Technical field
The present invention relates to Detection dynamic target technical field, particularly relate to a kind of dynamic target method for detecting based on video image.
Background technology
Conventional dynamic target method for detecting has optical flow method, frame differential method, background method of finite difference and the morphologic method of mathematics etc. Optical flow method when not knowing any information of scene in advance, can detect out independent mobile, but most optical flow method calculation of complex is consuming time, it is difficult to meet the requirement of detection in real time; Background difference rule is difficult to obtain the image frame not containing moving target completely and carrys out initialize background, if the unexpected athletic meeting of static object existed in background in mobile or background causes the appearance of empty static object, and interference detection results; Frame differential method and mathematics morphologic method algorithm are simple, there is stronger adaptivity, can requirement of real time, but have the moving object detection tolerance range of the branch etc. waved not high in as even in uneven illumination in complex scene, to there is gray scale different multiple moving targets, background.
Summary of the invention
It is an object of the invention to solve, by a kind of dynamic target method for detecting based on video image, the problem that above background section is mentioned.
For reaching this object, the present invention by the following technical solutions:
Based on a dynamic target method for detecting for video image, it comprises the steps:
S101, the image of input is carried out filtering remove noise;
S102, utilize three adjacent two field pictures, intermediate frame carried out difference with adjacent frame respectively, and then result phase with;
S103, the noise spot removed in binary map picture, false target area, then the border in living target region is improved by the method for region growing.
Especially, described step S103 specifically comprises: scanned by binary map picture, and isolated bright spot in removal figure also stores the number of accurate target area, each target area area and boundary coordinate thereof; Set a threshold value, remove all target areas being less than threshold value, obtain the seed region of target growth; Carry out region merging technique by the method for region growing and fill leak.
First adjacent image frame is carried out difference and two value difference gray-scale map pictures by the dynamic target method for detecting based on video image that the present invention proposes, then scan binary map picture and remove spurious motion object and isolated bright spot, the information of mobile is finally improved by the way of region growing, complicated in background, to there is multiple mobile video can accurately detect out each moving target and effectively be filled with the cavity of target, mobile can be detected out more accurately compared with tradition frame differential method, the morphologic method of mathematics etc.
Accompanying drawing explanation
The dynamic target method for detecting schema based on video image that Fig. 1 provides for the embodiment of the present invention.
Embodiment
Below in conjunction with drawings and Examples, the invention will be further described. It should be appreciated that specific embodiment described herein is only for explaining the present invention, but not limitation of the invention. It also should be noted that, for convenience of description, illustrate only part related to the present invention in accompanying drawing and not all content, unless otherwise defined, all technology used herein are identical with the implication that the those skilled in the art belonging to the present invention understand usually with scientific terminology. The term used in the description of the invention herein is the object in order to describe specific embodiment, is not intended to be restriction the present invention. Term as used herein " and/or " comprise arbitrary and all combinations of one or more relevant Listed Items.
Please refer to shown in Fig. 1, the dynamic target method for detecting schema based on video image that Fig. 1 provides for the embodiment of the present invention.
In the present embodiment, dynamic target method for detecting based on video image specifically comprises the steps:
S101, the image of input is carried out filtering remove noise. Video image is after pre-treatment, and in figure, the effect of visualization of area-of-interest will improve, and is conducive to as follow-up target detect, feature extraction, target tracking provide high-quality image.
S102, utilize three adjacent two field pictures, intermediate frame carried out difference with adjacent frame respectively, and then result phase with. The shape profile of intermediate frame motion target can be detected out like this. In two values of difference image, the precision chosen directly affecting target segmentation of threshold value. Threshold value Ding get Tai Gao, it is easy to a large amount of targets is judged to background, on the contrary easily a large amount of backgrounds is judged to target again. The sampling rate of pick up camera is generally greater than every second 15 frames, so the change of adjacent two field picture is little. Therefore, in difference image, the gray-scale value of the pixel in non-region of variation and background is very little, and region of variation and moving target region gray-scale value are bigger. This shows in difference image, and the gray-scale value of different zones pixel obeys different distributions: in background area, the gray-scale value of pixel is subject to influence of noise and presents Gaussian distribution, and in foreground target region, the grey value profile of pixel is non-gaussian.
S103, the noise spot removed in binary map picture, false target area, then the border in living target region is improved by the method for region growing. Being scanned by binary map picture, isolated bright spot in removal figure also stores the number of accurate target area, each target area area and boundary coordinate thereof. After scanning entire image, although eliminating isolated bright spot, but the accurate target area obtained further comprises some noise region becoming block and false target area. Owing to target area and the noise block area of these falsenesses are all little, it is possible to according to the size of living target in experience and original image, set a threshold value, remove all target areas being less than threshold value, obtain the seed region of target growth. Carry out region merging technique by the method for region growing and fill leak.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, to those skilled in the art, the present invention can have various change and change. Any amendment of doing within all spirit in the present invention and principle, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (2)
1. the dynamic target method for detecting based on video image, it is characterised in that, comprise the steps:
S101, the image of input is carried out filtering remove noise;
S102, utilize three adjacent two field pictures, intermediate frame carried out difference with adjacent frame respectively, and then result phase with;
S103, the noise spot removed in binary map picture, false target area, then the border in living target region is improved by the method for region growing.
2. the dynamic target method for detecting based on video image according to claim 1, it is characterized in that, described step S103 specifically comprises: scanned by binary map picture, and isolated bright spot in removal figure also stores the number of accurate target area, each target area area and boundary coordinate thereof; Set a threshold value, remove all target areas being less than threshold value, obtain the seed region of target growth; Carry out region merging technique by the method for region growing and fill leak.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410632496.4A CN105654503A (en) | 2014-11-11 | 2014-11-11 | Dynamic target detection method based on video images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410632496.4A CN105654503A (en) | 2014-11-11 | 2014-11-11 | Dynamic target detection method based on video images |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105654503A true CN105654503A (en) | 2016-06-08 |
Family
ID=56482412
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410632496.4A Pending CN105654503A (en) | 2014-11-11 | 2014-11-11 | Dynamic target detection method based on video images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105654503A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108663026A (en) * | 2018-05-21 | 2018-10-16 | 湖南科技大学 | A kind of vibration measurement method |
| CN110288558A (en) * | 2019-06-26 | 2019-09-27 | 纳米视觉(成都)科技有限公司 | A kind of super depth image fusion method and terminal |
-
2014
- 2014-11-11 CN CN201410632496.4A patent/CN105654503A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108663026A (en) * | 2018-05-21 | 2018-10-16 | 湖南科技大学 | A kind of vibration measurement method |
| CN108663026B (en) * | 2018-05-21 | 2020-08-07 | 湖南科技大学 | Vibration measuring method |
| CN110288558A (en) * | 2019-06-26 | 2019-09-27 | 纳米视觉(成都)科技有限公司 | A kind of super depth image fusion method and terminal |
| CN110288558B (en) * | 2019-06-26 | 2021-08-31 | 福州鑫图光电有限公司 | Super-depth-of-field image fusion method and terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109636732B (en) | Hole repairing method of depth image and image processing device | |
| CN113781402B (en) | Method and device for detecting scratch defects on chip surface and computer equipment | |
| Liu et al. | Single image dehazing via large sky region segmentation and multiscale opening dark channel model | |
| CN108154520B (en) | A kind of moving target detecting method based on light stream and frame matching | |
| US20110211749A1 (en) | System And Method For Processing Video Using Depth Sensor Information | |
| Wang et al. | Moving target detection and tracking based on Pyramid Lucas-Kanade optical flow | |
| Sun | A new single-image fog removal algorithm based on physical model | |
| CN104835164A (en) | Method and device for processing binocular camera depth image | |
| Lian et al. | A novel method on moving-objects detection based on background subtraction and three frames differencing | |
| CN111046862B (en) | Character segmentation method, device and computer readable storage medium | |
| CN112364865B (en) | Method for detecting small moving target in complex scene | |
| Liu et al. | Texture filtering based physically plausible image dehazing | |
| JP2021086616A (en) | Method for extracting effective region of fisheye image based on random sampling consistency | |
| CN110930361B (en) | Method for detecting occlusion of virtual and real objects | |
| Chen et al. | A color-guided, region-adaptive and depth-selective unified framework for Kinect depth recovery | |
| CN105654503A (en) | Dynamic target detection method based on video images | |
| Hsiao et al. | Robust multiple objects tracking using image segmentation and trajectory estimation scheme in video frames | |
| CN106558069A (en) | A kind of method for tracking target and system based under video monitoring | |
| US20180260960A1 (en) | Method and apparatus for assisted object selection in video sequences | |
| Zhou et al. | Rain detection and removal of sequential images | |
| CN106447681B (en) | A kind of object segmentation methods of non-uniform severe motion degraded image | |
| Xu | Study on moving objects by video monitoring system of recognition and tracing scheme | |
| JP5838112B2 (en) | Method, program and apparatus for separating a plurality of subject areas | |
| Yuxi et al. | A new motion obstacle detection based monocular-vision algorithm | |
| RU2555238C1 (en) | Digital image segmentation method (versions) |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160608 |