Face occlusion removal for face recognition using the related face by structural similarity index measure and principal component analysis

Facial occlusions like sunglasses, masks, caps etc. have severe consequences when reconstructing the partially occluded regions of a facial picture. This paper proposes a novel hybrid machine learning approach for occlusion removal based on Structural Similarity Index Measure (SSIM) and Principal Component Analysis (PCA), called SSIM_PCA. The proposed system comprises two stages. In the first stage, a Face Similar Matrix (FSM) guided by the Structural Similarity Index Measure is generated to provide the necessary information to recover from the lost regions of the face image. The FSM generates Related Face (RF) images similar to the probe image. In the second stage, these RF images are considered as related information and used as input data to generate eigenspaces using PCA to reconstruct the occluded face region exploiting the relationship between the occluded region and related face images, which contain relevant data to recover from the occluded area. Experimental results with three standard datasets viz. Caspeal-R1, IMFDB, and FEI have proven that the proposed method works well under illumination changes and occlusion of facial images.

Removing Foreground Occlusions in Light Field using Micro-lens Dynamic Filter

Foreground occlusion removal task aims to automatically detect and remove foreground occlusions and recover background objects. Since for Light Fields (LFs), background objects occluded in some views may be seen in other views, the foreground occlusion removal task for LFs is easy to achieve. In this paper, we propose a learning-based method combining ‘seeking’ and ‘generating’ to recover occluded background. Specifically, the micro-lens dynamic filters are proposed to ‘seek’ occluded background points in shifted micro-lens images and remove occlusions using angular information. The shifted images are then combined to further ‘generate’ background regions to supplement more background details using spatial information. By fully exploring the angular and spatial information in LFs, the dense and complex occlusions can be easily removed. Quantitative and qualitative experimental results show that our method outperforms other state-of-the-arts methods by a large margin.

Non-linear Sorenson–Dice Exemplar Image Inpainting Based Bayes Probability for Occlusion Removal in Remote Traffic Control

Occlusion removal is a significant problem to be resolved in a remote traffic control system to enhance road safety. However, the conventional techniques do not recognize traffic signs well due to the vehicles are occluded. Besides occlusion removal was not performed in existing techniques with a less amount of time. In order to overcome such limitations, Non-linear Gaussian Bilateral Filtered Sorenson–Dice Exemplar Image Inpainting Based Bayes Conditional Probability (NGBFSEII-BCP) Method is proposed. Initially, a number of remote sensing images are taken as input from Highway Traffic Dataset. Then, the NGBFSEII-BCP method applies the Non-Linear Gaussian Bilateral Filtering (NGBF) algorithm for removing the noise pixels in input images. After preprocessing, the NGBFSEII-BCP method is used to remove the occlusion in the input images. Finally, NGBFSEII-BCP Method applies Bayes conditional probability to find operation status and thereby gets higher road safety using remote sensing images. The technique conducts the simulation evaluation using metrics such as peak signal to noise ratio, computational time, and detection accuracy. The simulation result illustrates that the NGBFSEII-BCP Method increases the detection accuracy by 20% and reduces the computation time by 32% as compared to state-of-the-art works.