China 
Africa 
Explore chapters and articles related to this topic
Fundamental Methods in Image Processing
Published in Ling Guan, Yifeng He, Sun-Yuan Kung, Multimedia Image and Video Processing, 2012
April Khademi, Anastasios N. Venetsanopoulos, Alan R. Moody, Sridhar Krishnan
To this end, the current section discusses an edge-preserving smoothing filter, which retains the important image details, while smoothing in the flat regions. Such filters have been gaining more attention lately, as they combat the downfalls of traditional filters. The most popular edge-preserving smoothing algorithm to date is the bilateral filter [33], where filtering is completed via convolution with a kernel that is the product of two Gaussians:
Singular value decomposition-based anisotropic diffusion for fusion of infrared and visible images
Published in International Journal of Image and Data Fusion, 2019
Let and be infrared and visible images, respectively, and let all images be co-registered. These images are passed through the edge-preserving smoothing anisotropic diffusion process for obtaining the base layers denoted as and of the infrared and visible images, respectively, where = and =. The detail layers and are obtained by subtracting the base layers from the source images.
Multi-stage guided-filter for SAR and optical satellites images fusion using Curvelet and Gram Schmidt transforms for maritime surveillance
Published in International Journal of Image and Data Fusion, 2023
Tarek M. Ghoniemy, Mahmoud M. Hammad, A. S. Amein, Tarek A. Mahmoud
Since the published patent in He et al. (2013), guided filter has gained its importance in image processing field. It is a fast, good visual quality, easy and effective image fusion method. This fusion method is characterised by edge preserving smoothing property and does not suffer from artefacts like bilateral filter, and it has a local linear transformation of the guidance image (Meng et al. 2016, Ma et al. 2019). Motivated by the success in preserving spatial details, guided filter is extended to multi-stages as the input of each current stage depends on the output of the previous stage (Yang et al. 2016). Guided filter can be represented by the following equation:
Prediction of transverse crack multiplication of CFRP cross-ply laminates under tension-tension fatigue load
Published in Advanced Composite Materials, 2023
Youzou Kitagawa, Masahiro Arai, Akinori Yoshimura, Keita Goto
In order to quantitatively evaluate the difference in microscopic damage with the maximum fatigue load, the interfacial debonding ratio, which is the ratio of the interfacial debonding area to the transverse crack area, was investigated by image processing. The image processing procedure for the interfacial debonding ratio evaluation is briefly explained below; however, the full methodology is found in one of our previous studies [28]. We applied the edge-preserving smoothing process twice on the 8-bit gray images, using a bilateral filter [29].A binarization process was applied to separate the transverse crack from the images, and also to extract carbon fibers from the images.An edge-detection process using sobel filter [30] was employed to detect the crack surface and outlines of the carbon fibers.Thinning process based on Zhang–Suen thinning algorithm [31] was applied to the crack surface images, and we obtained the pixels that expressed the crack surface as a total of the interfacial debonding and matrix damage.We applied the AND process on the images of crack surfaces and outlines of fibers.We then obtained the interfacial debonding pixels, and calculated the interfacial debonding ratio by dividing the number of interfacial debonding pixels by the crack surface pixels.