CN108509926A - A kind of building extracting method based on two-way color notation conversion space - Google Patents
A kind of building extracting method based on two-way color notation conversion space Download PDFInfo
- Publication number
- CN108509926A CN108509926A CN201810307834.5A CN201810307834A CN108509926A CN 108509926 A CN108509926 A CN 108509926A CN 201810307834 A CN201810307834 A CN 201810307834A CN 108509926 A CN108509926 A CN 108509926A
- Authority
- CN
- China
- Prior art keywords
- component
- enhancing
- building
- value
- true
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/176—Urban or other man-made structures
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to a kind of building extracting methods based on two-way color notation conversion space.Include the following steps：Step 1, multi-spectrum remote sensing image is inputted；Step 2, RGB color turns LUV color spaces；Step 3, LUV is decomposed；Step 4, the enhancing respectively in extraction component L, U and V is really gathered；Step 5, LUV color spaces turn RGB color；Step 6, image segmentation；Step 7, it post-processes；Step 8, result is exported.The building in multi-spectrum remote sensing image can be accurately extracted, the update of building in the geographical basis information database of city is can be applied to.
Description
Technical field
The present invention relates to a kind of remote sensing image process field, specifically a kind of building based on two-way color notation conversion space
Object extracting method.
Background technology
Building is one of main geographic element in city, is the important content of various city thematic maps, studies building
Extraction is of great significance to integrated survey urban geographic information environment.It is quick with high-resolution remote sensing image acquiring technology
Development, the processing of remote sensing image, analysis and application have a better data source, digital product then have more extensively, deeper into
Application.Computer image processing technology, pattern-recognition, artificial intelligence etc. all obtain different degrees of progress, for height
Effective information in effect ground extraction huge image data provides possibility.But the extraction of building information than other information such as road,
The acquisition of water body is much more difficult, and main cause is as follows：
(1) data source is mainly two-dimensional remote sensing image, in most cases lacks direct three-dimensional data；
(2) different remote sensing image Chang Yinwei spectral regions, resolution ratio, the several picture of sensor and image-forming condition etc.
The difference of factor and have larger difference；
(3) its appearance for being showed of different types of building and grain details etc. are ever-changing, show remote sensing
Widely different on image, unified building model library is difficult to set up, this makes automatically extracting for information become extremely difficult；
(4) complexity of scene residing for building, when as relatively low such as contrast, house mutually block, the moon of building itself
Shadow and the shade etc. for being in other atural objects, so thinking that the building for automatically extracting sharpness of border from background is more tired
It is difficult.
Invention content
The present invention provides a kind of building extracting methods based on two-way color notation conversion space, can overcome current remote sensing shadow
The difficult problem of building extraction can detect in remote sensing image in conjunction with the advantages of two color spaces of RGB and LUV as in
Building target is not necessarily to manual intervention, high degree of automation.
Technical solution is used by target to realize the present invention：Method includes the following steps：
Step 1：Input width is W, highly for H and include R, the multi-spectrum remote sensing image I of tri- color components of G, Bin；
Step 2：By multi-spectrum remote sensing image IinColor space LUV is converted to from color space RGB；
Step 3：Color space LUV in step 2 is decomposed into component L, component U and component V；
Step 4：The component L of color space LUV is converted into true set (T with following formulaL), indefinite set (XL) and it is false
Gather (PL)：
In formula (1), TL(x, y) is true set (TL) in pixel (x, y) value, XL(x, y) is indefinite set (XL) in
The value of pixel (x, y), PL(x, y) is false set (PL) in pixel (x, y) value；Int (x, y) is pixel (x, y)
Intensity value,For centered on pixel (x, y), the intensity of all pixels point is average in the window that the length of side is w × w
Value,For the maximum value of average strength in component L,For the minimum value of average strength in component L；dif(x,y)
For the strength offsets amount of pixel (x, y),difgBe strength offsets amount threshold value, and
difg>difs；
Step 5：It is calculated separately with following formula and really gathers (T in step 4L), indefinite set (XL) and false set (PL) letter
Entropy is ceased, is denoted as E (T respectivelyL)、E(XL) and E (PL)：
Step 6：Merge the comentropy E (T in step 5 with following formulaL)、E(XL) and E (PL), obtain the information of component L
Entropy E (L)：
E (L)=E (TL)+E(XL)+E(PL)； (3)
Step 7：To the true set (T in step 4L), indefinite set (XL) and false set (PL) enhancing operation is carried out, specifically
Formula is：
In formula (4), TEN L、PEN LAnd XEN LEnhance true set in respectively component L, enhance false set and enhance indefinite set,
K is enhancing coefficient and can be sought by lower formula：
ME is image IinMaximum informational entropy, value log2 W×H；
Step 8：With the indefinite set X of enhancing in following formula judgment step 7EN LComentropy E (XEN L) whether stablize：
(E(XEN L)-E(XL))/E(XL)<δ, (6)
When formula (6) are set up, comentropy E (XEN L) stablize, 9 are entered step, otherwise TL=TEN L, XL=XEN L, PL=PEN L,
Enter step 5；
Step 9：The true set T of extraction enhancingEN L；
Step 10：The method that step 4 arrives step 9 is reused, extracts the enhancing very set T in component U respectivelyEN UWith point
Measure the enhancing very set T in VEN V；
Step 11：It will the true set T of enhancingEN L, enhance true set TEN UWith the true set T of enhancingEN VThe LUV color spaces of composition
RGB color is converted to, image I is obtainedT；
Step 12：To image ITIt is split, extraction cutting object set SEG；
Step 13：Rectangular degree and length-width ratio is used to extract building as constraints；
Step 14：Output extraction building result.
Dividing method in the step 12 uses Region growing segmentation method.
The beneficial effects of the invention are as follows：The building in multi-spectrum remote sensing image can be accurately extracted, city is can be applied to
The update of building in area's geographical basis information database.
Description of the drawings
Fig. 1 is the overall process flow figure of the present invention；
Fig. 2 is the flow chart for extracting the enhancing in component L in the present invention and really gathering.
Specific implementation mode
Detailed description of the present invention specific implementation mode below in conjunction with the accompanying drawings.
Fig. 1 is the overall process flow figure of the present invention, as shown in Figure 1：
In step 101, input width is W, highly for H and include R, the multi-spectrum remote sensing image of tri- color components of G, B
Iin。
In step 102, by the multi-spectrum remote sensing image I in step 101inColor space is converted to from color space RGB
LUV。
In step 103, the color space LUV in step 102 is decomposed into component L, component U and component V.
In step 104, the enhancing very set T in component L, component U and component V is extracted respectivelyEN L, enhance true set TEN UWith
The true set T of enhancingEN V。
In step 105, by the true set T of enhancingEN L, enhance true set TEN UWith the true set T of enhancingEN VThe LUV colors of composition are empty
Between be converted to RGB color, obtain image IT。
In step 106, using Region growing segmentation method to image ITIt is split, extraction cutting object set SEG.
In step 107, rectangular degree and length-width ratio is used to extract building as constraints.
In step 108, output extraction building result.
Fig. 2 is the flow chart for extracting the enhancing in component L in the present invention and really gathering, the i.e. increasing that step 104 in Fig. 1 includes
Strong true set TEN LFlow is extracted, as shown in Figure 2：
In step 201, input component L.
In step 202, the component L of color space LUV is converted into true set (T with following formulaL), indefinite set (XL)
Gather (P with vacationL)：
In formula (7), TL(x, y) is true set (TL) in pixel (x, y) value, XL(x, y) is indefinite set (XL) in
The value of pixel (x, y), PL(x, y) is false set (PL) in pixel (x, y) value, int (x, y) be pixel (x, y)
Intensity value,For centered on pixel (x, y), the intensity of all pixels point is average in the window that the length of side is w × w
Value,For the maximum value of average strength in component L,For the minimum value of average strength in component L, dif (x, y)
For the strength offsets amount of pixel (x, y),difgBe strength offsets amount threshold value, and
difg>difs。
In step 203, is calculated separately with following formula and really gather (T in step 202L), indefinite set (XL) and false set
(PL) comentropy, be denoted as E (T respectivelyL)、E(XL) and E (PL)：
In step 204, merge the comentropy E (T in step 203 with following formulaL)、E(XL) and E (PL), obtain component L
Comentropy E (L)：
E (L)=E (TL)+E(XL)+E(PL)。 (9)
In step 205, to the true set (T in step 202L), indefinite set (XL) and false set (PL) carry out enhancing behaviour
Make, specific formula is：
In formula (10), TEN L、PEN LAnd XEN LEnhance true set in respectively component L, enhance false set and enhances indefinite collection
It closes, k is enhancing coefficient and can be sought by lower formula：
ME is image IinMaximum informational entropy, value log2 W×H。
In step 206, with the indefinite set X of enhancing in following formula judgment step 205EN LComentropy E (XEN L) whether
Stablize：
(E(XEN L)-E(XL))/E(XL)<δ, (12)
When formula (12) are set up, comentropy E (XEN L) stablize, 207 are entered step, otherwise TL=TEN L, XL=XEN L, PL=
PEN L, enter step 203.
Really gather (T in the enhancing of step 207, output component LEN L)。
Using the process flow of Fig. 2, when step 201 inputs component U, the enhancing of extractable component U is really gathered
(TEN U), when step 201 inputs component V, (T is really gathered in the enhancing of extractable component UEN V)。
Claims (2)
1. a kind of building extracting method based on two-way color notation conversion space, it is characterised in that include the following steps：
Step 1：Input width is W, highly for H and include R, the multi-spectrum remote sensing image I of tri- color components of G, Bin；
Step 2：By multi-spectrum remote sensing image IinColor space LUV is converted to from color space RGB；
Step 3：Color space LUV in step 2 is decomposed into component L, component U and component V；
Step 4：The component L of color space LUV is converted into true set (T with following formulaL), indefinite set (XL) and false set
(PL：)
In formula (1), TL(x, y) is true set (TL) in pixel (x, y) value, XL(x, y) is indefinite set (XL) in pixel
The value of point (x, y), PL(x, y) is false set (PL) in pixel (x, y) value, int (x, y) be pixel (x, y) it is strong
Angle value,For centered on pixel (x, y), the average strength of all pixels point in the window that the length of side is w × w,For the maximum value of average strength in component L,For the minimum value of average strength in component L, dif (x, y) is picture
The strength offsets amount of vegetarian refreshments (x, y),difgBe strength offsets amount threshold value, and difg
>difs；
Step 5：It is calculated separately with following formula and really gathers (T in step 4L), indefinite set (XL) and false set (PL) information
Entropy is denoted as E (T respectivelyL)、E(XL) and E (PL)：
Step 6：Merge the comentropy E (T in step 5 with following formulaL)、E(XL) and E (PL), obtain the comentropy E of component L
(L)：
E (L)=E (TL)+E(XL)+E(PL)； (3)
Step 7：To the true set (T in step 4L), indefinite set (XL) and false set (PL) carry out enhancing operation, specific formula
For：
In formula (4), TEN L、PEN LAnd XEN LEnhance true set in respectively component L, enhance false set and enhance indefinite set, k is
Enhance coefficient and can be sought by lower formula：
ME is image IinMaximum informational entropy, value log2 W×H；
Step 8：With the indefinite set X of enhancing in following formula judgment step 7EN LComentropy E (XEN L) whether stablize：
(E(XEN L)-E(XL))/E(XL)<δ, (6)
When formula (6) are set up, comentropy E (XEN L) stablize, 9 are entered step, otherwise TL=TEN L, XL=XEN L, PL=PEN L, enter
Step 5；
Step 9：The true set T of extraction enhancingEN L；
Step 10：The method that step 4 arrives step 9 is reused, extracts the enhancing very set T in component U respectivelyEN UIn component V
Enhancing very set TEN V；
Step 11：It will the true set T of enhancingEN L, enhance true set TEN UWith the true set T of enhancingEN VThe LUV color space conversions of composition
For RGB color, image I is obtainedT；
Step 12：To image ITIt is split, extraction cutting object set SEG；
Step 13：Rectangular degree and length-width ratio is used to extract building as constraints；
Step 14：Output extraction building result.
2. a kind of building extracting method based on two-way color notation conversion space according to claim 1, it is characterised in that
Dividing method in step 12 uses Region growing segmentation method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810307834.5A CN108509926B (en) | 2018-04-08 | 2018-04-08 | Building extraction method based on bidirectional color space transformation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810307834.5A CN108509926B (en) | 2018-04-08 | 2018-04-08 | Building extraction method based on bidirectional color space transformation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108509926A true CN108509926A (en) | 2018-09-07 |
| CN108509926B CN108509926B (en) | 2021-06-01 |
Family
ID=63381209
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810307834.5A Expired - Fee Related CN108509926B (en) | 2018-04-08 | 2018-04-08 | Building extraction method based on bidirectional color space transformation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108509926B (en) |
Citations (3)
| 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 |
| CN107169946A (en) * | 2017-04-26 | 2017-09-15 | 西北工业大学 | Image interfusion method based on non-negative sparse matrix Yu hypersphere color transformation |
| CN107527007A (en) * | 2016-06-20 | 2017-12-29 | 戴尔菲技术公司 | For detecting the image processing system of perpetual object |
-
2018
- 2018-04-08 CN CN201810307834.5A patent/CN108509926B/en not_active Expired - Fee Related
Patent Citations (3)
| 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 |
| CN107527007A (en) * | 2016-06-20 | 2017-12-29 | 戴尔菲技术公司 | For detecting the image processing system of perpetual object |
| CN107169946A (en) * | 2017-04-26 | 2017-09-15 | 西北工业大学 | Image interfusion method based on non-negative sparse matrix Yu hypersphere color transformation |
Non-Patent Citations (4)
| Title |
|---|
| ALIREZA RAHIMZADEGANASL 等: "AUTOMATIC BUILDING DETECTION BASED ON CIE LUV COLOR SPACE USING VERY HIGH RESOLUTION PLEIADES IMAGES", 《IEEE》 * |
| WENZAO SHI 等: "Building extraction from panchromatic high-resolution remotely sensed imagery based on potential histogram and neighborhood Total variation", 《EARTH SCI INFORM》 * |
| 陈凯强: "高分辨率遥感图像建筑物提取方法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 顾明 等: "基于颜色空间转换的交通图像增强算法", 《仪器仪表学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108509926B (en) | 2021-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Spatially and temporally weighted regression: A novel method to produce continuous cloud-free Landsat imagery | |
| Dutta et al. | Growth of Dehradun city: An application of linear spectral unmixing (LSU) technique using multi-temporal landsat satellite data sets | |
| Cai et al. | Study on shadow detection method on high resolution remote sensing image based on HIS space transformation and NDVI index | |
| CN103871072B (en) | Orthography based on project digital elevation model inlays line extraction method | |
| Congalton | Remote sensing: an overview | |
| Huang et al. | A multiscale urban complexity index based on 3D wavelet transform for spectral–spatial feature extraction and classification: an evaluation on the 8-channel WorldView-2 imagery | |
| Wang et al. | Color matching simulation of ocean landscape decoration pattern based on visual communication | |
| CN109635715B (en) | Remote sensing image building extraction method | |
| Jiang et al. | Semi-automatic building extraction from high resolution imagery based on segmentation | |
| CN107977968B (en) | Building layered detection method based on building shadow information mining | |
| Sari et al. | Quality analysis of single tree object with obia and vegetation index from lapan surveillance aircraft multispectral data in urban area | |
| CN105761250A (en) | Building extraction method based on fuzzy scene segmentation | |
| CN108491826B (en) | Automatic extraction method of remote sensing image building | |
| Tran et al. | Classification of image matching point clouds over an urban area | |
| CN108509926A (en) | A kind of building extracting method based on two-way color notation conversion space | |
| Cubillas et al. | The application of support vector machine (SVM) using cielab color model, color intensity and color constancy as features for ortho image classification of Benthic Habitats in Hinatuan, Surigao del sur, Philippines | |
| Le Bris et al. | Cnn semantic segmentation to retrieve past land cover out of historical orthoimages and dsm: first experiments | |
| CN107392208B (en) | Object spectral feature extraction method based on spectral space mapping and purification | |
| CN105740825B (en) | It is a kind of for mixing the large format remote sensing image building extracting method of scene | |
| Caccetta et al. | Fine-scale monitoring of complex environments using remotely sensed aerial, satellite, and other spatial data | |
| CN108596088A (en) | A kind of building analyte detection method for panchromatic remote sensing image | |
| Motayyeb et al. | Enhancing contrast of images to improve geometric accuracy of a UAV photogrammetry project | |
| Viseur | Automated methods for fully exploring and interpreting LIDAR data points | |
| H. Trauth et al. | Image processing | |
| Du et al. | A new multi-feature approach to object-oriented change detection based on fuzzy classification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for 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: 20210601 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |