Hyperspectral image feature representation method based on deep joint sparse-collaborative representation

• Main Authors: LIU YILIU, CONG YUHUA, XING CHANGDA, WANG ZHISHENG, DUAN CHAOWEI, WANG MEILING
• Format: Patent
• Language: Chinese; English
• Published: 28.01.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses a hyperspectral image feature representation method based on deep joint sparse-collaborative representation. The method comprises the following steps: firstly, considering that each pixel of a hyperspectral image contains nonlinear information, redundancy exists among frequency spectrums of the pixels, and very strong correlation exists among similar pixel samples; secondly, providing a hyperspectral image feature representation method based on deep joint sparse-collaborative representation, so that significant information in pixel samples and correlation information between the samples can be represented at the same time, and deep nonlinear characteristics of the hyperspectral image can be extracted; and finally, designing an alternating iteration algorithm to solve the hyperspectral image feature representation method of deep joint sparse-collaborative representation to obtain a hyperspectral image feature representation form. According to the invention, nonlinear mapping is carried