CN103514599A - Image optimum segmentation dimension selecting method based on neighborhood total variation - Google Patents
Image optimum segmentation dimension selecting method based on neighborhood total variation Download PDFInfo
- Publication number
- CN103514599A CN103514599A CN201310386335.7A CN201310386335A CN103514599A CN 103514599 A CN103514599 A CN 103514599A CN 201310386335 A CN201310386335 A CN 201310386335A CN 103514599 A CN103514599 A CN 103514599A
- Authority
- CN
- China
- Prior art keywords
- image
- image segmentation
- segmentation result
- index
- result images
- 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.)
- Granted
Links
Images
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image optimum segmentation dimension selecting method based on a neighborhood total variation. The image optimum segmentation dimension selecting method comprises the first step of segmenting an image with different segmentation dimensions to obtain a series of image segmentation result images, the second step of calculating homogeneity indexes of interiors of targets in the image segmentation result images under different dimensions according to the homogeneity evaluation method, the third step of calculating heterogeneity indexes among the image segmentation result image targets under different dimensions according to the heterogeneity evaluation method, and the fourth step of normalizing the homogeneity indexes and the heterogeneity indexes and assigning the weight of the heterogeneity indexes. The segmentation dimension corresponding to the minimum of the comprehensive evaluation indexes is the optimum segmentation dimension of the image. The image segmentation result is evaluated by adopting the neighborhood total variation, the homogeneity of the interior of the target of the image segmentation and the heterogeneity among the targets are comprehensively taken into consideration, an objective selection method is provided for the image optimum segmentation dimension, and the improvement of the segmentation effect and the follow-up processing of the image are facilitated.
Description
Technical field
The present invention relates to the technical field of the best segmental scale of How to choose remote sensing image, relate in particular to a kind of image optimum based on neighborhood total variation and cut apart scale selection method.
Background technology
Image Segmentation is the committed step of object-oriented image processing, and the quality cut apart can directly affect the subsequent treatment effect of image, wherein cuts apart in the segmentation effect that is chosen in image of yardstick and has very important effect.After best segmental scale is defined as image and utilizes this yardstick to cut apart, ground class can be expressed by one or several object, and object size and ground object target size approach, and the gray scale of imaged object inside is more even, and the gray difference between object is larger.The selection of the current best segmental scale about image has several different methods, as object and neighborhood absolute mean difference variance ratio (Ratio of Mean Difference to Neighbors to Standard Deviation, RMAS), vector distance method, but these two kinds of methods are biased to some extent, can not consider more all sidedly the problem of image less divided or over-segmentation.The segmentation effect that Espindola analyzes image by spatial autocorrelation carries out image optimum cuts apart scale selection, but choice criteria wherein can not embody the optimality of Image Segmentation well.
Summary of the invention
The object of the invention is to solve a difficult problem for Image Segmentation scale selection, provide a kind of method based on neighborhood total variation to carry out best segmental scale selection.
Image optimum based on neighborhood total variation is cut apart scale selection method and is comprised the steps:
1) image is cut apart with the different yardsticks of cutting apart, obtained a series of Image Segmentation result images;
2) according to homogeney evaluation method, calculate the homogeney index H that each cuts apart object inside in the Image Segmentation result images under yardstick;
3) according to heterogeneous evaluation method, calculate each and cut apart the heterogeneity index I between object in the Image Segmentation result images under yardstick;
4) by homogeney index H and heterogeneity index I normalization, specify the weight of heterogeneity index I, calculate the comprehensive evaluation index F (H of the Image Segmentation result images under each yardstick, I), obtain Image Segmentation comprehensive evaluation index F (H, I) minimum value, the minimum value of comprehensive evaluation index F (H, I) is corresponding cuts apart the best segmental scale that yardstick is image.
Described step 2) comprising:
(1) obtain total number n of cutting object in each Image Segmentation result images;
(2) set the area a of k cutting object in each Image Segmentation result images
kwith gray standard deviation v
k;
(3) by following formula, try to achieve the homogeney index H of all cutting objects inside in each Image Segmentation result images,
Described step 3) comprises:
(1) by following formula, try to achieve the mould of the gradient g of raw video f (x, y) | g (f (x, y)) |,
(2) setting cutting object border radius neighborhood scope radius in each Image Segmentation result images is r, and neighborhood scope is
obtain neighborhood scope
on number of pixels
(3) by following formula, try to achieve neighborhood scope in each Image Segmentation result images
the mould of upper image gradient | g (f (x, y)) | integration,
(4) by following formula, try to achieve the heterogeneity index I between imaged object in each Image Segmentation result images,
Described step 4) comprises:
(1) press two formulas by the homogeney index H of all Image Segmentation result images and heterogeneity index I normalization respectively,
(2) specify the weight ρ of heterogeneity index I, press homogeney index H and the heterogeneity index I of following formula combined imaging segmentation result image, try to achieve the Image Segmentation evaluation number F (H, I) of each Image Segmentation result images,
F(H,I)=(1-ρ)F(H)+ρF(I);
(3) ask the minimum value of Image Segmentation comprehensive evaluation index F (H, I), the minimum value of comprehensive evaluation index F (H, I) is corresponding cuts apart the best segmental scale that yardstick is image.
The beneficial effect that the present invention compared with prior art has:
(1) considered the homogeney of Image Segmentation object inside and the heterogeneity between object, comprehensively objective to the evaluation analysis of Image Segmentation quality;
(2) homogeney index calculation method of the present invention is subject to the restriction of standard deviation, area and object sum simultaneously, can consider on the whole from image the homogeney of image, the over-segmentation in can objective measurement Image Segmentation and before cut apart phenomenon;
(3) heterogeneity index computing application neighborhood total variation of the present invention, can significantly embody heterogeneity between object and the optimality of Image Segmentation yardstick, for Image Segmentation provides best segmental scale choice criteria.
Accompanying drawing explanation:
Fig. 1 is the segmentation result figure of cutting apart under 10 different scales of the present invention;
Fig. 2 is Image Segmentation yardstick evaluation number of the present invention and the curve map of cutting apart yardstick.
embodiment:
Image optimum based on neighborhood total variation is cut apart scale selection method and is comprised the steps:
1) image is cut apart with the different yardsticks of cutting apart, obtained a series of Image Segmentation result images;
2) according to homogeney evaluation method, calculate the homogeney index H that each cuts apart object inside in the Image Segmentation result images under yardstick;
3) according to heterogeneous evaluation method, calculate each and cut apart the heterogeneity index I between object in the Image Segmentation result images under yardstick;
4) by homogeney index H and heterogeneity index I normalization, specify the weight of heterogeneity index I, calculate the comprehensive evaluation index F (H of the Image Segmentation result images under each yardstick, I), obtain Image Segmentation comprehensive evaluation index F (H, I) minimum value, the minimum value of comprehensive evaluation index F (H, I) is corresponding cuts apart the best segmental scale that yardstick is image.
Described step 2) comprising:
(1) obtain total number n of cutting object in each Image Segmentation result images;
(2) set the area a of k cutting object in each Image Segmentation result images
kwith gray standard deviation v
k;
(3) by following formula, try to achieve the homogeney index H of all cutting objects inside in each Image Segmentation result images,
Described step 3) comprises:
(1) by following formula, try to achieve the mould of the gradient g of raw video f (x, y) | g (f (x, y)) |,
(2) setting cutting object border radius neighborhood scope radius in each Image Segmentation result images is r, and neighborhood scope is
obtain neighborhood scope
on number of pixels
(3) by following formula, try to achieve neighborhood scope in each Image Segmentation result images
the mould of upper image gradient | g (f (x, y)) | integration,
(4) by following formula, try to achieve the heterogeneity index I between imaged object in each Image Segmentation result images,
Described step 4) comprises:
(1) press two formulas by the homogeney index H of all Image Segmentation result images and heterogeneity index I normalization respectively,
(2) specify the weight ρ of heterogeneity index I, press homogeney index H and the heterogeneity index I of following formula combined imaging segmentation result image, try to achieve the Image Segmentation evaluation number F (H, I) of each Image Segmentation result images,
F(H,I)=(1-ρ)F(H)+ρF(I);
(3) ask the minimum value of Image Segmentation comprehensive evaluation index F (H, I), the minimum value of comprehensive evaluation index F (H, I) is corresponding cuts apart the best segmental scale that yardstick is image.
Embodiment:
The first step, selecting somewhere resolution is that the fast bird image of 2.4 meters is as experimental data, the eCognition software of applying German DEFINIENS company carries out multi-scale division 10 times to image, cut apart at every turn cut apart yardstick be set to respectively 50,60,70 ..., 140, other parameter remains unchanged, and the segmentation result obtaining as shown in Figure 1;
Second step, obtains total number n of the cutting object in each Image Segmentation result images, the area a of each cutting object k
kwith gray standard deviation v
k, by following formula, try to achieve the homogeney index H of all cutting objects inside in each Image Segmentation result images,
The 3rd step, first by formula
try to achieve the mould of the gradient g of raw video f (x, y) | g (f (x, y)) |, then set radius neighborhood scope radius r=3 in cutting object border in each Image Segmentation result images, establish neighborhood scope and be
obtain neighborhood scope
on number of pixels
by formula
try to achieve neighborhood scope in each Image Segmentation result images
the mould of upper image gradient | g (f (x, y)) | integration, finally by following formula, try to achieve the heterogeneity index I between imaged object in each Image Segmentation result images,
The 4th step, by homogeney index H and heterogeneity index I normalization, specify the weight ρ of heterogeneity index I, calculate the Image Segmentation comprehensive evaluation index F (H, I) of the Image Segmentation result images under each yardstick, statistics Image Segmentation comprehensive evaluation index F (H, I) and cut apart the relation of yardstick s, obtain curve map as shown in Figure 2, the yardstick of cutting apart that in figure, Image Segmentation comprehensive evaluation index minimum value is corresponding is 130, i.e. 130 best segmental scales that are this image.
Claims (4)
1. the image optimum based on neighborhood total variation is cut apart a scale selection method, it is characterized in that comprising the steps:
1) image is cut apart with the different yardsticks of cutting apart, obtained a series of Image Segmentation result images;
2) according to homogeney evaluation method, calculate the homogeney index H that each cuts apart object inside in the Image Segmentation result images under yardstick;
3) according to heterogeneous evaluation method, calculate each and cut apart the heterogeneity index I between object in the Image Segmentation result images under yardstick;
4) by homogeney index H and heterogeneity index I normalization, specify the weight of heterogeneity index I, calculate the comprehensive evaluation index F (H of the Image Segmentation result images under each yardstick, I), obtain Image Segmentation comprehensive evaluation index F (H, I) minimum value, the minimum value of comprehensive evaluation index F (H, I) is corresponding cuts apart the best segmental scale that yardstick is image.
2. a kind of image optimum based on neighborhood total variation according to claim 1 is cut apart scale selection method, it is characterized in that described step 2) comprising:
(1) obtain total number n of cutting object in each Image Segmentation result images;
(2) set the area a of k cutting object in each Image Segmentation result images
kwith gray standard deviation v
k;
(3) by following formula, try to achieve the homogeney index H of all cutting objects inside in each Image Segmentation result images,
3. a kind of image optimum based on neighborhood total variation according to claim 1 is cut apart scale selection method, it is characterized in that described step 3) comprises:
(1) by following formula, try to achieve the mould of the gradient g of raw video f (x, y) | g (f (x, y)) |,
(2) setting cutting object border radius neighborhood scope radius in each Image Segmentation result images is r, and neighborhood scope is
obtain neighborhood scope
on number of pixels
(3) by following formula, try to achieve neighborhood scope in each Image Segmentation result images
the mould of upper image gradient | g (f (x, y)) | integration,
(4) by following formula, try to achieve the heterogeneity index I between imaged object in each Image Segmentation result images,
4. a kind of image optimum based on neighborhood total variation according to claim 1 is cut apart scale selection method, it is characterized in that described step 4) comprises:
(1) press two formulas by the homogeney index H of all Image Segmentation result images and heterogeneity index I normalization respectively,
(2) specify the weight ρ of heterogeneity index I, press homogeney index H and the heterogeneity index I of following formula combined imaging segmentation result image, try to achieve the Image Segmentation evaluation number F (H, I) of each Image Segmentation result images,
F(H,I)=(1-ρ)F(H)+ρF(I);
(3) ask the minimum value of Image Segmentation comprehensive evaluation index F (H, I), the minimum value of comprehensive evaluation index F (H, I) is corresponding cuts apart the best segmental scale that yardstick is image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310386335.7A CN103514599B (en) | 2013-08-30 | 2013-08-30 | A kind of segmentation of the image optimum based on neighborhood total variation scale selection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310386335.7A CN103514599B (en) | 2013-08-30 | 2013-08-30 | A kind of segmentation of the image optimum based on neighborhood total variation scale selection method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103514599A true CN103514599A (en) | 2014-01-15 |
CN103514599B CN103514599B (en) | 2016-02-24 |
Family
ID=49897276
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310386335.7A Active CN103514599B (en) | 2013-08-30 | 2013-08-30 | A kind of segmentation of the image optimum based on neighborhood total variation scale selection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103514599B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106529430A (en) * | 2016-10-31 | 2017-03-22 | 武汉大学 | High spatial resolution image residential area extraction method based on characteristics of spatial correlation and heterogeneity |
CN107067405A (en) * | 2017-03-30 | 2017-08-18 | 河海大学 | Based on the preferred Remote Sensing Image Segmentation of yardstick |
CN107194942A (en) * | 2017-03-27 | 2017-09-22 | 广州地理研究所 | It is a kind of to determine the method that image classification splits yardstick threshold value |
CN109816668A (en) * | 2019-01-22 | 2019-05-28 | 中国科学院地理科学与资源研究所 | The non-supervisory segmentation evaluation method and apparatus of remote sensing image |
CN113850822A (en) * | 2021-09-18 | 2021-12-28 | 四川大学 | Automatic slope unit dividing method based on confluence segmentation |
CN114119645A (en) * | 2021-11-25 | 2022-03-01 | 推想医疗科技股份有限公司 | Method, system, device and medium for determining image segmentation quality |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101126812A (en) * | 2007-09-27 | 2008-02-20 | 武汉大学 | High resolution ratio remote-sensing image division and classification and variety detection integration method |
US20130071003A1 (en) * | 2011-06-22 | 2013-03-21 | University Of Florida | System and device for characterizing cells |
-
2013
- 2013-08-30 CN CN201310386335.7A patent/CN103514599B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101126812A (en) * | 2007-09-27 | 2008-02-20 | 武汉大学 | High resolution ratio remote-sensing image division and classification and variety detection integration method |
US20130071003A1 (en) * | 2011-06-22 | 2013-03-21 | University Of Florida | System and device for characterizing cells |
Non-Patent Citations (3)
Title |
---|
付树军 等: "基于各向异性扩散方程的超声图像去噪与边缘增强", 《电子学报》, vol. 33, no. 7, 31 July 2005 (2005-07-31) * |
何敏 等: "面向对象的最优分割尺度计算模型", 《大地测量与地球动力学》, vol. 29, no. 1, 28 February 2009 (2009-02-28) * |
李书晓,常红星: "基于总变分和形态学的航空图像道路检测算法", 《计算机学报》, vol. 30, no. 12, 31 December 2007 (2007-12-31) * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106529430A (en) * | 2016-10-31 | 2017-03-22 | 武汉大学 | High spatial resolution image residential area extraction method based on characteristics of spatial correlation and heterogeneity |
CN107194942A (en) * | 2017-03-27 | 2017-09-22 | 广州地理研究所 | It is a kind of to determine the method that image classification splits yardstick threshold value |
CN107067405A (en) * | 2017-03-30 | 2017-08-18 | 河海大学 | Based on the preferred Remote Sensing Image Segmentation of yardstick |
CN107067405B (en) * | 2017-03-30 | 2020-04-03 | 河海大学 | Remote sensing image segmentation method based on scale optimization |
CN109816668A (en) * | 2019-01-22 | 2019-05-28 | 中国科学院地理科学与资源研究所 | The non-supervisory segmentation evaluation method and apparatus of remote sensing image |
CN113850822A (en) * | 2021-09-18 | 2021-12-28 | 四川大学 | Automatic slope unit dividing method based on confluence segmentation |
CN113850822B (en) * | 2021-09-18 | 2023-04-25 | 四川大学 | Automatic slope unit dividing method based on confluence division |
CN114119645A (en) * | 2021-11-25 | 2022-03-01 | 推想医疗科技股份有限公司 | Method, system, device and medium for determining image segmentation quality |
Also Published As
Publication number | Publication date |
---|---|
CN103514599B (en) | 2016-02-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110378196B (en) | Road visual detection method combining laser point cloud data | |
CN103514599B (en) | A kind of segmentation of the image optimum based on neighborhood total variation scale selection method | |
CN109658398B (en) | Part surface defect identification and evaluation method based on three-dimensional measurement point cloud | |
CN106651872B (en) | Pavement crack identification method and system based on Prewitt operator | |
CN108491757B (en) | Optical remote sensing image target detection method based on multi-scale feature learning | |
Zhou et al. | Real-time kiwifruit detection in orchard using deep learning on Android™ smartphones for yield estimation | |
Ferraz et al. | Lidar detection of individual tree size in tropical forests | |
US20180088228A1 (en) | Obstacle detection method and apparatus for vehicle-mounted radar system | |
CN110263717B (en) | Method for determining land utilization category of street view image | |
JP2021504816A (en) | Bone age evaluation and height prediction model, its system and its prediction method | |
CN104463164B (en) | It is a kind of based on umbrella frame method and crown height than trees canopy structure information extracting method | |
CN105930772A (en) | City impervious surface extraction method based on fusion of SAR image and optical remote sensing image | |
CN103870834B (en) | Method for searching for sliding window based on layered segmentation | |
CN107909018B (en) | Stable multi-mode remote sensing image matching method and system | |
Liu et al. | Estimating potato above-ground biomass by using integrated unmanned aerial system-based optical, structural, and textural canopy measurements | |
CN113763350B (en) | Glue line detection method and device, glue line detection equipment and storage medium | |
CN110910445B (en) | Object size detection method, device, detection equipment and storage medium | |
CN102651124A (en) | Image fusion method based on redundant dictionary sparse representation and evaluation index | |
CN102855485A (en) | Automatic wheat earing detection method | |
CN105205816A (en) | Method for extracting high-resolution SAR image building zone through multi-feature weighted fusion | |
CN111079807B (en) | Ground object classification method and device | |
Maurya et al. | Development of fusion approach for estimation of vegetation fraction cover with drone and sentinel-2 data | |
CN106529548A (en) | Sub-pixel level multi-scale Harris corner detection algorithm | |
CN116188943A (en) | Solar radio spectrum burst information detection method and device | |
Xu et al. | Vegetation information extraction in karst area based on UAV remote sensing in visible light band |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |