Matching Method of Airborne Down-view Heterogeneous Images Based on Region Division
A technology of area division and matching method, which is applied in the field of image matching, can solve the problem of low matching accuracy of airborne images, achieve the effect of weak inter-class area similarity, high target matching accuracy, and expand the application range
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment approach 1
[0041] Specific implementation mode one: combine Figure 8 This embodiment will be described. The region division-based airborne down-view heterogeneous image matching method given in this embodiment is used for such as figure 1 The target image shown and the complex and heterogeneous airborne real-time image, the method first uses the standard deviation STD of the direction histogram as a parameter to determine the texture characteristics of the target image; if the target image is a rich texture image, use the image segmentation based The image matching method completes the airborne down-view image localization process, which uses image segmentation to generate mask images for different regions of the image. In these regions, the improved SIFT image matching method is used to match each region, and the orientation histogram-based The evaluation function obtains the optimal matching area as the matching result; for non-rich texture images, the image matching method based on a...
specific Embodiment approach 2
[0066] Specific embodiment two: the difference between this embodiment and specific embodiment one is that in step (3)
[0067] If the target image is a rich texture image, use the SIFT feature matching method to match each mask real-time image area with all mask target image areas; if the target image is a non-rich texture image, use SIFT feature matching In the method, in the process of consistent matching of each mask real-time image area and mask target image area, corner points (Corner) are added as feature key points.
[0068] The traditional SIFT algorithm includes four parts: scale space extremum detection, key point location, direction setting and key point description operator determination. The improved SIFT image matching method combines corner key points and SIFT key points together to form the key point set of the algorithm, uses the corner area as a mask to generate key points in the corner area, and adds corner points as the SIFT image of feature key points Th...
specific Embodiment approach 3
[0100] Specific embodiment three: the difference between this embodiment and specific embodiment one is that the evaluation function based on the direction histogram described in step (4) is used to evaluate the matching result, and the optimal matching area is selected as the specific matching result. The process includes:
[0101] (4.1) Use the Bhattacharyachian distance BD as the matching similarity measurement coefficient of the two regional histograms for matching;
[0102] (4.2) If the BD is greater than the threshold T, it is determined that the matching is successful; otherwise, the BD is less than or equal to the threshold T, and the matching fails;
[0103] (4.3) Among the successfully matched areas, select the area corresponding to the largest BD value as the optimal matching area.
[0104] Other steps and parameters are the same as those in Embodiment 1 or 2.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More