CN104794456A - Remote-sensing image building detecting method based on colors and shadows - Google Patents
Remote-sensing image building detecting method based on colors and shadows Download PDFInfo
- Publication number
- CN104794456A CN104794456A CN201510221017.4A CN201510221017A CN104794456A CN 104794456 A CN104794456 A CN 104794456A CN 201510221017 A CN201510221017 A CN 201510221017A CN 104794456 A CN104794456 A CN 104794456A
- Authority
- CN
- China
- Prior art keywords
- sensing image
- buildings
- remote sensing
- color
- method based
- 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
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a remote-sensing image building detecting method based on colors and shadows. The remote-sensing image building detecting method based on the colors and the shadows comprises the following steps of (1) extracting buildings with red roofs by using a color unchanging feature; (2) extracting shadow regions by using a shadow unchanging feature; (3) determining a sun illumination direction and estimating the positions of buildings without red roofs; and (4) fitting a region with the buildings without the red roofs by using a rectangle. The problem that in the traditional method of detecting buildings by using a shadow feature, priori knowledge of the solar illumination direction is required is solved, the detection accuracy and the detection efficiency of buildings are improved, and a complete automation effect is achieved.
Description
Technical field
The present invention relates to a kind of remote sensing image process field, is a kind of remote sensing image building detection method based on color and shade specifically.
Background technology
From remote sensing image, automatically extract man-made features quite complicated, within a very long time, people extract from remote sensing image that buildings pays attention to scientific research, have larger gap from practicality.In actual production, as remote sensing mapping production field, to identification, the extraction of building information, mainly by operator artificial cognition, artificial pointwise, to gather piecemeal, labour intensity is large, intricate operation, bad measurement repeatability, operating efficiency are low.Along with remote sensing image data amount increases, buildings upgrades pace of change to be accelerated, and urgently gets up gradually to the demand of remote sensing image intelligence interpretation system.Adopt automatic or semiautomatic fashion extraction building information, can greatly reduce data acquisition time, reply GIS data gathers difficult bottleneck; The problems such as artificial image data point finite sum subjective factor impact can be overcome simultaneously, improve the quality that GIS enters database data.
Summary of the invention
The invention provides a kind of remote sensing image building detection method based on color and shade, can estimate to shine upon direction and the position of accurate location buildings, without the need to the support in side information data storehouse, automaticity is high.
The technical scheme adopted for realizing target of the present invention is: method comprises the following steps:
Step 1: utilize color invariant features to extract region-of-interest to input multi-spectrum remote sensing image image1;
Step 2: utilize shade invariant features to extract shadow region to input multi-spectrum remote sensing image image1;
Step 3: utilize shade to determine to shine upon direction and estimate the position of the buildings on non-red roof;
Step 4: with the non-red roof building object area of rectangle fitting.
Described region-of-interest refers to the buildings on red roof.
Described color invariant features is defined by the red band R in input multi-spectrum remote sensing image image1 and green band G, and formula is as follows:
Can be obtained by formula, for the atural object of redness,
for on the occasion of, and with intensity have nothing to do, similarly, for the atural object of green,
for negative value, carry out Automatic thresholding by OTSU and split.
Described shade invariant features is defined by the blue wave band B in input multi-spectrum remote sensing image image1 and green band G, and formula is as follows:
Can be obtained by formula, for shadow region,
for on the occasion of, and with intensity have nothing to do, similarly, for the atural object of green,
for negative value, carry out Automatic thresholding by OTSU and split.
The method that described determination shines upon direction is: suppose to shine upon direction from buildings center directional phantom regional center, make buildings centre coordinate be (x
b, y
b), shadow region centre coordinate is (x
s, y
s), then the illumination direction theta of the sun is:
And determine the coordinate quadrant residing for θ by the nearly step of following formula:
The method of the position of the buildings on the non-red roof of described estimation adopts following formula:
x
e=x
s+dcosθ
y
e=y
s+dsinθ
Wherein, (x
e, y
e) centre coordinate of buildings for estimating, (x
s, y
s) centre coordinate of shadow region for extracting, θ is for shining upon direction, and d is average buildings size in image.
The described method by rectangle fitting construction zone is: choose adjacent with shadow region, along shining upon the buildings of the rectangular area on direction as non-red roof.
The invention has the beneficial effects as follows: solve in classic method the problem utilizing shadow character detection buildings to need to shine upon the priori in direction, be conducive to the accuracy and efficiency improving buildings detection, reach full automatic effect.
Accompanying drawing explanation
Fig. 1 is overall process flow figure of the present invention.
Embodiment
The specific embodiment of the present invention is described in detail below in conjunction with accompanying drawing.
In step 101, the pending remote sensing image of input is Quick bird multispectral image, is of a size of 1000 × 1000, and carries out the pre-service such as radiant correction and geometry correction.
In step 102, utilize color invariant features to be defined by the red band R in input multi-spectrum remote sensing image image1 and green band G, formula is as follows:
Carry out Automatic thresholding by OTSU and split, extracting the buildings on red roof, be designated as B
red.
In step 103, utilize shade invariant features to be defined by the blue wave band B in input multi-spectrum remote sensing image image1 and green band G, formula is as follows:
Carry out Automatic thresholding by OTSU and split, extracting shadow region, be designated as SH.
In step 104, according to the building B on the red roof extracted in step 102
redshadow region SH with extracting in step 103, supposes to shine upon direction from buildings center directional phantom regional center, makes the building B on red roof
redcentre coordinate is (x
b, y
b), shadow region SH centre coordinate is (x
s, y
s), then the illumination direction theta of the sun is:
And determine the coordinate quadrant residing for θ by the nearly step of following formula:
In step 105, the computing method for the position of the buildings on the non-red roof in the pending remote sensing image of input adopt following formula:
x
e=x
s+dcosθ
y
e=y
s+dsinθ
Wherein, (x
e, y
e) centre coordinate of buildings for estimating, (x
s, y
s) centre coordinate of shadow region SH for extracting in step 103, θ be calculate in step 104 shine upon direction, d is average buildings size in image, and value is 50.
In step 106, choose adjacent with shadow region, along the rectangular area shined upon on direction as non-red roof building thing B
nred.
In step 107, the red roof building thing B extracted in combining step 102
redwith the non-red roof building thing B that step 106 is extracted
nredexport as the owned building object detected.
Claims (7)
1., based on a remote sensing image building detection method for color and shade, it is characterized in that comprising the following steps:
Step 1: utilize color invariant features to extract region-of-interest to input multi-spectrum remote sensing image image1;
Step 2: utilize shade invariant features to extract shadow region to input multi-spectrum remote sensing image image1;
Step 3: utilize shade to determine to shine upon direction and estimate the position of the buildings on non-red roof;
Step 4: with the non-red roof building object area of rectangle fitting.
2. a kind of remote sensing image building detection method based on color and shade according to claim 1, is characterized in that region-of-interest refers to the buildings on red roof.
3. a kind of remote sensing image building detection method based on color and shade according to claim 1, it is characterized in that color invariant features is defined by the red band R in input multi-spectrum remote sensing image image1 and green band G, formula is as follows:
Can be obtained by formula, for the atural object of redness,
for on the occasion of, and with intensity have nothing to do, similarly, for the atural object of green,
for negative value, carry out Automatic thresholding by OTSU and split.
4. a kind of remote sensing image building detection method based on color and shade according to claim 1, it is characterized in that shade invariant features is defined by the blue wave band B in input multi-spectrum remote sensing image image1 and green band G, formula is as follows:
Can be obtained by formula, for shadow region,
for on the occasion of, and with intensity have nothing to do, similarly, for the atural object of green,
for negative value, carry out Automatic thresholding by OTSU and split.
5. a kind of remote sensing image building detection method based on color and shade according to claim 1, it is characterized in that the method determining to shine upon direction is: suppose to shine upon direction from buildings center directional phantom regional center, make buildings centre coordinate be (x
b, y
b), shadow region centre coordinate is (x
s, y
s), then the illumination direction theta of the sun is:
And determine the coordinate quadrant residing for θ by the nearly step of following formula:
6. a kind of remote sensing image building detection method based on color and shade according to claim 1, is characterized in that the method for the position of the buildings estimating non-red roof adopts following formula:
x
e=x
s+dcosθ
y
e=y
s+dsinθ
Wherein, (x
e, y
e) centre coordinate of buildings for estimating, (x
s, y
s) centre coordinate of shadow region for extracting, θ is for shining upon direction, and d is average buildings size in image.
7. a kind of remote sensing image building detection method based on color and shade according to claim 1, it is characterized in that by the method for rectangle fitting construction zone be: choose adjacent with shadow region, along shining upon the buildings of the rectangular area on direction as non-red roof.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510221017.4A CN104794456B (en) | 2015-05-04 | 2015-05-04 | A kind of remote sensing image building detection method based on color and shade |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510221017.4A CN104794456B (en) | 2015-05-04 | 2015-05-04 | A kind of remote sensing image building detection method based on color and shade |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104794456A true CN104794456A (en) | 2015-07-22 |
CN104794456B CN104794456B (en) | 2018-01-16 |
Family
ID=53559243
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510221017.4A Expired - Fee Related CN104794456B (en) | 2015-05-04 | 2015-05-04 | A kind of remote sensing image building detection method based on color and shade |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104794456B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105761250A (en) * | 2016-02-01 | 2016-07-13 | 福建师范大学 | Building extraction method based on fuzzy scene segmentation |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103279951A (en) * | 2013-05-13 | 2013-09-04 | 武汉理工大学 | Object-oriented remote sensing image building and shade extraction method of remote sensing image building |
US8577170B2 (en) * | 2011-09-15 | 2013-11-05 | Microsoft Corporation | Shadow detection in a single image |
-
2015
- 2015-05-04 CN CN201510221017.4A patent/CN104794456B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8577170B2 (en) * | 2011-09-15 | 2013-11-05 | Microsoft Corporation | Shadow detection in a single image |
CN103279951A (en) * | 2013-05-13 | 2013-09-04 | 武汉理工大学 | Object-oriented remote sensing image building and shade extraction method of remote sensing image building |
Non-Patent Citations (2)
Title |
---|
刘小洲等: "高分辨率遥感图像中建筑物阴影的提取", 《计算机仿真》 * |
徐秋红等: "一种基于颜色恒常性理论的城市高分辨率遥感影像阴影消除方法", 《理论研究》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105761250A (en) * | 2016-02-01 | 2016-07-13 | 福建师范大学 | Building extraction method based on fuzzy scene segmentation |
Also Published As
Publication number | Publication date |
---|---|
CN104794456B (en) | 2018-01-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020258899A1 (en) | Mountain landscape architecture extraction method and equipment based on high-resolution remote sensing image | |
Bao et al. | Field-based architectural traits characterisation of maize plant using time-of-flight 3D imaging | |
CN103914675B (en) | A kind of 2 D code of quick response matrix recognition methods of clothing | |
CN105956557B (en) | A kind of sequential remote sensing image cloud covered areas domain automatic testing method of object-oriented | |
CN103455813A (en) | Method for facula center positioning of CCD image measurement system | |
CN104036516B (en) | Camera calibration checkerboard image angular-point detection method based on symmetrical analysis | |
CN108089544B (en) | Trajectory generation method and control system for sole glue spraying robot | |
CN103955583B (en) | Method for determining threshold value of urban built-up area extracted through nighttime light data | |
CN106599890B (en) | digital instrument recognition algorithm for substation inspection robot | |
CN103400117B (en) | A kind of personnel in the pit's positioning and tracing method based on computer vision | |
CN105138990A (en) | Single-camera-based gesture convex hull detection and palm positioning method | |
Cai et al. | Study on shadow detection method on high resolution remote sensing image based on HIS space transformation and NDVI index | |
CN103440035A (en) | Gesture recognition system in three-dimensional space and recognition method thereof | |
CN102855621A (en) | Infrared and visible remote sensing image registration method based on salient region analysis | |
CN111881801B (en) | Newly-added construction land remote sensing monitoring method and equipment based on invariant detection strategy | |
CN101976114A (en) | System and method for realizing information interaction between computer and pen and paper based on camera | |
CN102592288A (en) | Method for matching pursuit of pedestrian target under illumination environment change condition | |
CN105590316A (en) | Object-oriented high-resolution remote-sensing image shadow extraction method | |
CN102930279A (en) | Image identification method for detecting product quantity | |
CN111339948A (en) | Automatic identification method for newly-added buildings of high-resolution remote sensing images | |
CN103942555A (en) | Method for detecting nitrogen content of plant through images | |
CN104966101A (en) | Solar cell classification method based on LabVIEW | |
CN103942809A (en) | Method for detecting joint fissures in rock images | |
CN110455201A (en) | Stalk plant height measurement method based on machine vision | |
CN108830896A (en) | A kind of remote sensing image water body patch extracting method based on full convolutional network and architectural shadow geometrical model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
EXSB | Decision made by sipo to initiate substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180116 |
|
CF01 | Termination of patent right due to non-payment of annual fee |