CN104392236B - A kind of non-scalability camera UAS image characteristic extracting method - Google Patents
A kind of non-scalability camera UAS image characteristic extracting method Download PDFInfo
- Publication number
- CN104392236B CN104392236B CN201410773969.2A CN201410773969A CN104392236B CN 104392236 B CN104392236 B CN 104392236B CN 201410773969 A CN201410773969 A CN 201410773969A CN 104392236 B CN104392236 B CN 104392236B
- Authority
- CN
- China
- Prior art keywords
- image
- uas
- gray
- carried out
- camera
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000012937 correction Methods 0.000 claims abstract description 11
- 239000011159 matrix material Substances 0.000 claims description 21
- 230000011218 segmentation Effects 0.000 claims description 11
- 238000004064 recycling Methods 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 abstract description 6
- 238000003709 image segmentation Methods 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to Digital Image Processing and field of Computer Graphics, particularly relate to a kind of non-scalability camera UAS image characteristic extracting method, specifically comprise the following steps: read non-measured type camera UAS image; Gray proces or intermediate value process are carried out to non-measured type camera UAS image, to carry out denoising, obtains gray level image; Gray level image is split, to obtain image zooming-out result; Boundary mergence in contiguous range is carried out to image zooming-out result, to obtain the merging image of smooth correction; Store and merge image.One provided by the invention non-scalability camera UAS image characteristic extracting method, feature extraction and correction can be carried out for non-scalability camera UAS image, image is strengthened, Image Segmentation Methods Based on Features and target identification, thus make image become truly available.
Description
Technical field
The present invention relates to a kind of non-scalability camera UAS image characteristic extracting method, belong to Digital Image Processing and field of Computer Graphics.
Background technology
At present, control UAS image a little with having accurate, utilize this image to carry out feature extraction excellent effect, but this unmanned plane and system thereof is expensive.
And it is limited to take without the common non-scalability camera unmanned plane of ground control point the image definition come, need a large amount of later stage corrections and feature extraction that image just can be made to become truly available, and existing feature extracting method usually cannot meet the requirement of Image Segmentation Methods Based on Features and target identification.
Summary of the invention
In order to solve the deficiencies in the prior art, the invention provides a kind of non-scalability camera UAS image characteristic extracting method, feature extraction and correction can be carried out for non-scalability camera UAS image, image is strengthened, Image Segmentation Methods Based on Features and target identification, thus make image become truly available.
The technical scheme that the present invention adopts for its technical matters of solution is: provide a kind of non-scalability camera UAS image characteristic extracting method, specifically comprise the following steps:
(1) the non-measured type camera UAS image that picture size is no more than 1GB is read;
(2) gray proces or intermediate value process are carried out to non-measured type camera UAS image, to carry out denoising, obtain gray level image;
(3) statistics is carried out to the gray-scale value of gray level image and obtain gray-scale statistical figure; When gray-scale statistical figure is tending towards normal distribution, adopt maximum between-cluster variance method computed segmentation threshold value, recycling binary segmentation method is split gray level image, namely obtains the image zooming-out result of black-and-white image; When gray-scale statistical figure has multimodal, then adopt kernel function matching Variance Method in space to obtain multi thresholds, recycle many-valued dividing method and gray level image is split, namely obtain the image zooming-out result of multistage grayscale image;
(4) boundary mergence in contiguous range is carried out to image zooming-out result, to obtain the merging image of smooth correction;
(5) merging image is stored.
In step (2), gray proces is carried out to non-measured type camera UAS image, specifically comprise the following steps: non-measured type camera UAS image is carried out colour gamut affined transformation, it is made to be converted to the binary matrix set type of digital picture, obtain the R access matrix component of non-measured type camera UAS image, G access matrix component and channel B matrix component, respectively gray processing process is carried out to three access matrix components, adopt the maximal value in three access matrix components as the respective value of new gray matrix again, obtain gray level image.
In step (2), intermediate value process is carried out to non-measured type camera UAS image, specifically comprise the following steps: for each object pixel in non-measured type camera UAS image, a template is arranged for each object pixel on image, described template comprises the adjacent pixels in the N*N contiguous range of object pixel, replace the value of object pixel with the mean value of pixels all in template, obtain the gray level image after intermediate value process; Described N is settings.
Step (4), in merging process, utilizes the smoothing process of linear filter method, finally obtains the merging image after smooth correction.
The present invention is based on the beneficial effect that its technical scheme has to be:
(1) one provided by the invention non-scalability camera UAS image characteristic extracting method, by carrying out gray proces or intermediate value process for non-scalability camera UAS image, denoising can be carried out, to improve the precision of feature extraction to fuzzy image;
(2) one provided by the invention non-scalability camera UAS image characteristic extracting method, different conditions for gray-scale map takes different dividing methods, pointed, segmentation efficiency and segmentation effect can be improved, and then improve the precision of feature extraction;
(3) one provided by the invention non-scalability camera UAS image characteristic extracting method is especially high for the non-scalability camera UAS image processing efficiency of size within 1GB;
(4) one provided by the invention non-scalability camera UAS image characteristic extracting method, taking smooth correction strategy, can strengthen image when merging, thus makes image become truly available.
Embodiment
Below in conjunction with embodiment, the invention will be further described.
The invention provides a kind of non-scalability camera UAS image characteristic extracting method, specifically comprise the following steps:
The technical scheme that the present invention adopts for its technical matters of solution is: provide a kind of non-scalability camera UAS image characteristic extracting method, specifically comprise the following steps:
(1) non-measured type camera UAS image is read; Described picture size is no more than 1GB.
(2) gray proces or intermediate value process are carried out to non-measured type camera UAS image, to carry out denoising, obtain gray level image:
If carry out gray proces to non-measured type camera UAS image, specifically comprise the following steps: non-measured type camera UAS image is carried out colour gamut affined transformation, it is made to be converted to the binary matrix set type of digital picture, obtain the R access matrix component of non-measured type camera UAS image, G access matrix component and channel B matrix component, respectively gray processing process is carried out to three access matrix components, adopt the maximal value in three access matrix components as the respective value of new gray matrix again, obtain gray level image.
If carry out intermediate value process to non-measured type camera UAS image, specifically comprise the following steps: for each object pixel in non-measured type camera UAS image, a template is arranged for each object pixel on image, described template comprises the adjacent pixels in the N*N contiguous range of object pixel, replace the value of object pixel with the mean value of pixels all in template, obtain the gray level image after intermediate value process.
(3) gray level image is split, to obtain image zooming-out result:
Statistics is carried out to the value of gray level image and obtains gray-scale statistical figure;
If gray-scale statistical figure is tending towards normal distribution, then adopt maximum between-cluster variance computed segmentation threshold value, recycling binary segmentation method is split gray level image, obtains black-and-white image, i.e. image zooming-out result; If gray-scale statistical figure has multimodal, then adopt kernel function matching Variance Method in space to obtain multi thresholds, recycle many-valued dividing method and gray level image is split, obtain multistage grayscale image, be i.e. image zooming-out result.
If gray-scale statistical figure is tending towards normal distribution, then adopt maximum between-cluster variance computed segmentation threshold value, recycling binary segmentation method is split gray level image, obtains black-and-white image, i.e. image zooming-out result; If gray-scale statistical figure has multimodal, then adopt kernel function matching Variance Method in space to obtain multi thresholds, recycle many-valued dividing method and gray level image is split, obtain multistage grayscale image, be i.e. image zooming-out result.
In this step, the multiple methods such as OTSU method, OSP-OTSU method, MeanShift method can be adopted to carry out Threshold segmentation.
(4) boundary mergence in contiguous range is carried out to image zooming-out result, to obtain the merging image of smooth correction: utilize the smoothing process of linear filter method, finally obtain the merging image after smooth correction.Namely refer on image to object pixel give a template, this template includes the adjacent pixels around it, such as, if 3*3 neighborhood, then using 8 pixels of the surrounding centered by object pixel as a Filtering Template, namely remove object pixel itself; 5*5 neighborhood then using 24 pixels around object pixel as a Filtering Template, then replace original pixel value with the mean value of the entire pixels in template.
(5) merging image is stored.
Claims (2)
1. a non-scalability camera UAS image characteristic extracting method, is characterized in that specifically comprising the following steps:
(1) the non-measured type camera UAS image that picture size is no more than 1GB is read;
(2) gray proces or intermediate value process are carried out to non-measured type camera UAS image, to carry out denoising, obtain gray level image; Gray proces is carried out to non-measured type camera UAS image, specifically comprise the following steps: non-measured type camera UAS image is carried out colour gamut affined transformation, it is made to be converted to the binary matrix set type of digital picture, obtain the R access matrix component of non-measured type camera UAS image, G access matrix component and channel B matrix component, respectively gray processing process is carried out to three access matrix components, adopt the maximal value in three access matrix components as the respective value of new gray matrix again, obtain gray level image;
Intermediate value process is carried out to non-measured type camera UAS image, specifically comprise the following steps: for each object pixel in non-measured type camera UAS image, a template is arranged for each object pixel on image, described template comprises the adjacent pixels in the N*N contiguous range of object pixel, replace the value of object pixel with the mean value of pixels all in template, obtain the gray level image after intermediate value process; Described N is settings;
(3) statistics is carried out to the gray-scale value of gray level image and obtain gray-scale statistical figure; When gray-scale statistical figure is tending towards normal distribution, adopt maximum between-cluster variance method computed segmentation threshold value, recycling binary segmentation method is split gray level image, namely obtains the image zooming-out result of black-and-white image; When gray-scale statistical figure has multimodal, then adopt kernel function matching Variance Method in space to obtain multi thresholds, recycle many-valued dividing method and gray level image is split, namely obtain the image zooming-out result of multistage grayscale image;
(4) boundary mergence in contiguous range is carried out to image zooming-out result, to obtain the merging image of smooth correction;
(5) merging image is stored.
2. non-scalability camera UAS image characteristic extracting method according to claim 1, is characterized in that: step (4), in merging process, utilizes the smoothing process of linear filter method, finally obtains the merging image after smooth correction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410773969.2A CN104392236B (en) | 2014-12-15 | 2014-12-15 | A kind of non-scalability camera UAS image characteristic extracting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410773969.2A CN104392236B (en) | 2014-12-15 | 2014-12-15 | A kind of non-scalability camera UAS image characteristic extracting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104392236A CN104392236A (en) | 2015-03-04 |
| CN104392236B true CN104392236B (en) | 2016-02-10 |
Family
ID=52610137
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410773969.2A Expired - Fee Related CN104392236B (en) | 2014-12-15 | 2014-12-15 | A kind of non-scalability camera UAS image characteristic extracting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104392236B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101877050A (en) * | 2009-11-10 | 2010-11-03 | 青岛海信网络科技股份有限公司 | Automatic extracting method for characters on license plate |
| CN101976437A (en) * | 2010-09-29 | 2011-02-16 | 中国资源卫星应用中心 | High-resolution remote sensing image variation detection method based on self-adaptive threshold division |
| CN103136735A (en) * | 2013-03-07 | 2013-06-05 | 中国科学院光电技术研究所 | Single image defogging method based on dual-scale dark channel |
-
2014
- 2014-12-15 CN CN201410773969.2A patent/CN104392236B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101877050A (en) * | 2009-11-10 | 2010-11-03 | 青岛海信网络科技股份有限公司 | Automatic extracting method for characters on license plate |
| CN101976437A (en) * | 2010-09-29 | 2011-02-16 | 中国资源卫星应用中心 | High-resolution remote sensing image variation detection method based on self-adaptive threshold division |
| CN103136735A (en) * | 2013-03-07 | 2013-06-05 | 中国科学院光电技术研究所 | Single image defogging method based on dual-scale dark channel |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104392236A (en) | 2015-03-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103914683A (en) | Gender identification method and system based on face image | |
| US9401027B2 (en) | Method and apparatus for scene segmentation from focal stack images | |
| CN102156868A (en) | Image binaryzation method and device | |
| CN106127817B (en) | A kind of image binaryzation method based on channel | |
| CN110969046B (en) | Face recognition method, face recognition device and computer-readable storage medium | |
| CN102509095B (en) | Number plate image preprocessing method | |
| CN107659799B (en) | Image pickup apparatus, image processing method, and storage medium | |
| CN105160682A (en) | Image edge detection method and apparatus | |
| CN108711160B (en) | Target segmentation method based on HSI (high speed input/output) enhanced model | |
| CN104021566A (en) | GrabCut algorithm-based automatic segmentation method of tongue diagnosis image | |
| CN104361335B (en) | A kind of processing method that black surround is automatically removed based on scan image | |
| CN111798470A (en) | Crop image entity segmentation method and system applied to intelligent agriculture | |
| CN103632153A (en) | Region-based image saliency map extracting method | |
| CN113902641A (en) | Data center hot area distinguishing method and system based on infrared image | |
| CN111145105A (en) | Image rapid defogging method and device, terminal and storage medium | |
| CN107256539B (en) | Image sharpening method based on local contrast | |
| CN104268845A (en) | Self-adaptive double local reinforcement method of extreme-value temperature difference short wave infrared image | |
| CN110766636A (en) | Intelligent image processing method for AI chip | |
| CN104182976A (en) | Field moving object fining extraction method | |
| CN110545414B (en) | Image sharpening method | |
| CN106952236B (en) | Fisheye lens shot image distortion correction method based on BP neural network | |
| CN110930358B (en) | Solar panel image processing method based on self-adaptive algorithm | |
| CN104392236B (en) | A kind of non-scalability camera UAS image characteristic extracting method | |
| CN110322479B (en) | Dual-core KCF target tracking method based on space-time significance | |
| CN110310274B (en) | Plant flower number detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C53 | Correction of patent of invention or patent application | ||
| CB03 | Change of inventor or designer information |
Inventor after: Chen Gang Inventor after: Zhao Ruyue Inventor after: Xu Wei Inventor after: Wu Peng Inventor before: Zhao Ruyue Inventor before: Chen Gang Inventor before: Xu Wei Inventor before: Wu Peng |
|
| COR | Change of bibliographic data |
Free format text: CORRECT: INVENTOR; FROM: ZHAO RUYUE CHEN GANG XU WEI WU PENG TO: CHEN GANG ZHAO RUYUE XU WEI WU PENG |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160210 Termination date: 20171215 |