CN102005044A - Method for detecting single-frequency interference based on edge direction histogram principle - Google Patents
Method for detecting single-frequency interference based on edge direction histogram principle Download PDFInfo
- Publication number
- CN102005044A CN102005044A CN 201010539187 CN201010539187A CN102005044A CN 102005044 A CN102005044 A CN 102005044A CN 201010539187 CN201010539187 CN 201010539187 CN 201010539187 A CN201010539187 A CN 201010539187A CN 102005044 A CN102005044 A CN 102005044A
- Authority
- CN
- China
- Prior art keywords
- edge
- histogram
- frequency
- image
- frequency interference
- 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
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method for detecting single-frequency interference based on an edge direction histogram and application thereof to a video fault diagnosis system. By the method, the single-frequency interference in images can be effectively and accurately detected so as to realize intelligent video fault diagnosis and analysis.
Description
Technical field
The invention belongs to computer vision field, particularly a kind of single-frequency interference detection method based on edge orientation histogram, and this method is in the video Application in Fault Diagnosis.
Technical background
Along with the rapid propelling in China of safe city engineering, the capital construction of video monitoring system begins to take shape, but simultaneously, the video monitoring system scale is increasing, monitored picture quantity is also more and more, and monitored picture quantity is also more and more, only depends on human eye that the fault picture is investigated one by one, efficient is very low, and therefore intelligentized video fault diagnosis importance highlights day by day.
The single-frequency interference is by the striated noise of the generation of generations such as imageing sensor, transmission channel, decoding processing, greatly reduced picture quality, thereby the single-frequency Interference Detection belongs to the diagnosis item of a key in the video fault diagnosis.Single-frequency interference detection method relatively more commonly used at present is filter method, histogram of gradients method and direction histogram method.Because common single-frequency interference detection method dependence color histogram or spatial color histogram are as extracting feature, fully differentiate between images target signature and single-frequency undesired signal under the situation of image color quality difference, industry is demanded urgently a kind of applied widely, the single-frequency interference detection method that accuracy is high.
Summary of the invention
The objective of the invention is at existing video fault diagnosis system, existence can't well detect the problem that single-frequency is disturbed, and proposes a kind of single-frequency interference detection method based on edge orientation histogram.
In order to realize goal of the invention, the technical scheme of employing is as follows:
The process flow diagram of single-frequency Interference Detection algorithm as shown in Figure 1.This flow process is at first carried out color space transformation to image, is transformed into the HSV color space; Extracting its H component is used for detecting and judgement; The H component is carried out rim detection, extract its edge; Follow the first order difference of difference edge calculation image x direction and y direction; Calculate its edge gradient directional image then, edge gradient direction image is carried out statistics with histogram; Ultimate analysis edge gradient figure histogram checks that the ratio whether certain direction scope accounts for is bigger than normal, if exist then key diagram looks like to exist single-frequency to disturb.
This algorithm based on condition select the H component statistics marginal information of image, judge that with marginal information single-frequency disturbs.Can be described below with the detailed process of this algorithm based on the single-frequency Interference Detection:
Extract the H component in the HSV color space;
This algorithm extracts the H component and carries out rim detection, and edge image is asked for the edge gradient direction histogram, ultimate analysis edge gradient direction histogram, and it is higher to compare the classic method accuracy rate.
Description of drawings
Fig. 1 is an architectural schematic of the present invention;
Fig. 2 is that RGB of the present invention changes HSV computing function synoptic diagram;
Fig. 3 is a gradient direction computing function synoptic diagram of the present invention.
Embodiment
Function of the present invention is based on up-to-date OpenCV storehouse.OpenCV is writing a Chinese character in simplified form of " Open Source Computer Vision Library ", is the Intel computer vision storehouse of increasing income.It is made of a series of C functions and a spot of C++ class, is a lot of general-purpose algorithms that can realize Flame Image Process and computer vision aspect, can be used to common problem in the process computer vision field, wherein is mainly concerned with the content of the following aspects:
(1) Color Space Conversion-color space transformation;
(2) Edge Detection-rim detection;
(3) Gradient in The Histogram-gradient orientation histogram;
(4) Histogram Statistics-statistics with histogram;
In the present invention, can carry out color space to image by function cvCvtColor and change, convert image to the HSV color space by rgb color space, conversion formula as shown in Figure 2.
Function cvSplit is used for the HSV color space of image is decomposed, and extracts the H color component.
Function cvCanny is used for rim detection, uses the Canny edge detection method that image is carried out rim detection.
Function cvSobel is used for the first order difference image of the x direction and the y direction of computed image, and the Sobel operator combines the level and smooth and differential of Gaussian, its as a result the paired pulses noise certain robustness is arranged.。
Function cvDiv is used to calculate the ratio of x direction difference diagram and y direction difference diagram, calculates the arc-tangent value of this ratio then, obtains gradient direction, and formula as shown in Figure 3.
Function cvCalcHist is used for the histogram of edge calculation gradient direction figure, obtains the statistics with histogram result.
Judge histogram, check that the ratio whether certain direction scope accounts for is bigger than normal, if exist then key diagram looks like to exist single-frequency to disturb.
Claims (4)
1. a method of disturbing based on edge orientation histogram principle detection single-frequency is characterized in that the edge of image direction is carried out statistics with histogram, and the parallel lines proportion is to judge whether image disturbed by single-frequency in the statistic histogram;
2. a method of disturbing based on edge orientation histogram principle detection single-frequency is characterized in that the function library based on OpenCV;
3. a method of disturbing based on edge orientation histogram principle detection single-frequency is characterized in that calculating the edge of image directional diagram based on the edge directional information of image x axle and y axle, counts edge orientation histogram according to edge orientation map at last;
4. one kind is detected the method that single-frequency is disturbed based on the edge orientation histogram principle, it is characterized in that the proportion that accounts for based on parallel lines in the statistics edge orientation histogram, judges during greater than threshold value when this proportion the single-frequency interference has taken place.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010539187 CN102005044A (en) | 2010-11-10 | 2010-11-10 | Method for detecting single-frequency interference based on edge direction histogram principle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010539187 CN102005044A (en) | 2010-11-10 | 2010-11-10 | Method for detecting single-frequency interference based on edge direction histogram principle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102005044A true CN102005044A (en) | 2011-04-06 |
Family
ID=43812383
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010539187 Pending CN102005044A (en) | 2010-11-10 | 2010-11-10 | Method for detecting single-frequency interference based on edge direction histogram principle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102005044A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030128396A1 (en) * | 2002-01-07 | 2003-07-10 | Xerox Corporation | Image type classification using edge features |
| US20060110036A1 (en) * | 2004-11-23 | 2006-05-25 | Hui Luo | Automated radiograph classification using anatomy information |
| CN101196996A (en) * | 2007-12-29 | 2008-06-11 | 北京中星微电子有限公司 | Image detection method and device |
| CN101211411A (en) * | 2007-12-21 | 2008-07-02 | 北京中星微电子有限公司 | Human body detection process and device |
-
2010
- 2010-11-10 CN CN 201010539187 patent/CN102005044A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030128396A1 (en) * | 2002-01-07 | 2003-07-10 | Xerox Corporation | Image type classification using edge features |
| US20060110036A1 (en) * | 2004-11-23 | 2006-05-25 | Hui Luo | Automated radiograph classification using anatomy information |
| CN101211411A (en) * | 2007-12-21 | 2008-07-02 | 北京中星微电子有限公司 | Human body detection process and device |
| CN101196996A (en) * | 2007-12-29 | 2008-06-11 | 北京中星微电子有限公司 | Image detection method and device |
Non-Patent Citations (1)
| Title |
|---|
| 《Pattern Recognition,2002,Proceedings. 16th International Conference on》 20021231 Yi Li et al. Consistent Line Clusters for Building Recognition in CBIR 第953页 1-4 , 2 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102043950B (en) | Vehicle outline recognition method based on canny operator and marginal point statistic | |
| CN105453153B (en) | Traffic lights detects | |
| CN105787929A (en) | Skin rash point extraction method based on spot detection | |
| EP1868162A3 (en) | Unusual action detection based on image analysis involving CHLAC feature extraction | |
| CN104198497B (en) | Surface defect detection method based on visual saliency map and support vector machine | |
| JP2019505866A (en) | Passerby head identification method and system | |
| CN107392885A (en) | A kind of method for detecting infrared puniness target of view-based access control model contrast mechanism | |
| CA2552875A1 (en) | System and method for filtering a medical image | |
| WO2023279966A1 (en) | Multi-lane-line detection method and apparatus, and detection device | |
| CN103679146A (en) | Safety belt intelligent detection method based on high-pass filter and Hough conversion | |
| CN103996051A (en) | Method for automatically detecting abnormal behaviors of video moving object based on change of movement features | |
| CN104318211A (en) | Anti-shielding face tracking method | |
| CN102881106A (en) | Dual-detection forest fire identification system through thermal imaging video and identification method thereof | |
| CN103902985A (en) | High-robustness real-time lane detection algorithm based on ROI | |
| CN104616006A (en) | Surveillance video oriented bearded face detection method | |
| CN105787912B (en) | Classification-based step type edge sub-pixel positioning method | |
| CN103996045A (en) | Multi-feature fused smoke identification method based on videos | |
| CN103646242B (en) | Extended target tracking based on maximum stable extremal region feature | |
| CN105046316B (en) | A kind of two-way pedestrian counting method of laser returned based on Gaussian process | |
| CN106097725A (en) | A kind of vehicle classification flow rate testing methods extracted based on behavioral characteristics | |
| CN104008396A (en) | In and out people flow statistical method based on people head color and shape features | |
| CN103034864A (en) | Video banner identification method based on color threshold and corner detection | |
| CN102005044A (en) | Method for detecting single-frequency interference based on edge direction histogram principle | |
| Sanmiguel et al. | Pixel-based colour contrast for abandoned and stolen object discrimination in video surveillance | |
| CN108010076B (en) | End face appearance modeling method for intensive industrial bar image detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110406 |