Abstract
Image matching forms an essential means of data association for computer vision, photogrammetry and remote sensing. The quality of image matching is heavily dependent on image details and naturalness. However, complex illuminations, denoting extreme and changing illuminations, are inevitable in real scenarios, and seriously deteriorate image matching performance due to their significant influence on the image naturalness and details. In this paper, a spatial-frequency domain associated image-optimization method, comprising two main models, is specially designed for improving image matching with complex illuminations. First, an adaptive luminance equalization is implemented in the spatial domain to reduce radiometric variations, instead of removing all illumination components. Second, a frequency domain analysis-based feature-enhancement model is proposed to enhance image features while preserving image naturalness and restraining over-enhancement. The proposed method associates the advantages of the spatial and frequency domain analyses to complete illumination equalization, feature enhancement and naturalness preservation, and thus acquiring the optimized images that are robust to the complex illuminations. More importantly, our method is generic and can be embedded in most image-matching schemes to improve image matching. The proposed method was evaluated on two different datasets and compared with four other state-of-the-art methods. The experimental results indicate that the proposed method outperforms other methods under complex illuminations, in both matching performances and practical applications such as structure from motion and multi-view stereo.
Highlights
Image matching is used to establish the correspondence between two or more images of the same scene taken from different viewpoints by the same or different sensors
We propose an adaptive method to equalize the differences in pixel in pixel intensities caused by the complex illuminations, and reduce the effect of the complex intensities caused by the complex illuminations, and reduce the effect of the complex illuminations on image features, which is the main mission
6489 and the mean reprojection error (RE) are calculated to evaluate the effect of sparse reconstruction, where the success rate of image localization (SRIL) is defined as SRIL = (NSL/NL) × 100%; here, NSL is the number of successful image localizations and NL is the number of images calculated
Summary
Image matching is used to establish the correspondence between two or more images of the same scene taken from different viewpoints by the same or different sensors. The second class is the changing illumination, in which the overlap regions of images with the same reflectance present obvious differences in the irradiance. Under this condition, color of objects and their textures might present significant variations [24]. Numerous feature detectors or descriptors are available in the literature [25,26,27,28,29] The majority of these methods can be used for illumination-invariant image matching, but show low applicability in practical situations [30]. A spatial-frequency domain associated image-optimization method, which optimizes the image naturalness and details, is proposed to improve image matching in the presence of the complex illuminations
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
