Collaborative Representation for Hyperspectral Image Classification and Detection
Qian Du, Chiranjibi Shah, Hongjun Su, Wei Li · 2024
Collaborative representation (CR) has been used in hyperspectral image analysis, such as classification and detection. Its basic idea is to represent a testing sample using labeled ones with a constraint of ℓ 2 -norm minimization of coefficient vector, which allows all the samples to have equal chance to participate in the representation. Variants of CR-based classifier (CRC) and CR-based detector (CRD) have been proposed to improve the performance of the original versions. The closed-form solutions offer implementation simplicity and computational efficiency. The performance is comparable or even better than other sophisticated machine learning approaches.