Linear Representation-Based Methods for Image Classification: A Survey

Abstract

In recent years, linear representation-based methods have been widely researched and applied in the image classification field. Generally speaking, there are three steps within linear representation-based classification (LRC) algorithms. The first step is coding, which uses all training samples to represent the test sample in a linear combination. The second step is subspace approximation, where residuals between the test sample and the linear combination of each class are calculated. The third step is classification, which assigns the class label to the minimum class-specific residual. We classify the LRC methods into six categories: 1) linear representation-based classification methods with norm minimizations, 2) linear representationbased classification methods with constraints, 3) linear representation-based classification methods with feature spaces, 4) linear representation-based classification methods with structural information, 5) linear representation with subspace learning, and 6) linear representation in semi-supervised learning and unsupervised learning. The purpose of this paper is to: 1) make an accurate and clear definition of the linear representation-based method, 2) provide a categorization and a comprehensive survey of the existing linear representation-based classification methods for image classification, 3) Summarize the main applications of linear representation-based methods, 4) provide extensive classification results and a discussion of the linear representation-based methods. Furthermore, this paper summarizes specific applications of the linear representation-based methods. Particularly, we performed extensive experiments to compare twelve linear representation-based classification methods on seven image classification datasets.

Enhanced Multi-Object Detection via the Integration of PSO, Kalman Filtering, and CNN Compressive Sensing

“…The outcome designates that IMM-CKF is extremely vigorous than the CKF when the space object experiences a movement. Shantaiya et al [5] has proposed multiple objects tracking from the video series using optical flow algorithm. This algorithm uses Kalman filtering to detect and track the moving objects in each frame, which helps in identifying moving objects with occlusion, blurred objects and so on.…”

“…Another is inconsequential templates, which include noisy pixels. The pixels that make up object patterns are sparse manner described in a low-dimensional sub space, resulting in a path of non-zero entrances in the sparse medium, which represents the location of the pixel in a precise image frame [5]. However, the object model with thin depiction grieves from heavy obstruction and cannot account for things that come and disappear over time.…”

Application of improved virtual sample and sparse representation in face recognition

“…The method obtains excellent and robust recognition results in face recognition. Zhou et al [15] defined the linear representation-based classification (LRC) algorithm accurately and clearly, and analysed and summarised it according to categories, providing extensive classification results and discussions. The 'Symmetric Face' combines the features of the original image to represent the face object, reducing the influence of face appearance changes, thus improving the accuracy of recognition [16].…”