Early tumor tissue recognition system based on hyperspectral image

The invention discloses an early tumor tissue recognition system based on a hyperspectral image, and relates to the technical field of hyperspectral image processing. Comprises: an ambient light capturing module configured to capture and pre-process ambient light features; the feature extraction module is configured to extract a tumor tissue feature region of the hyperspectral image and mark the tumor tissue feature region; the image processing module is configured to eliminate the influence of the ambient light, and perform preprocessing and spectral band optimization on the tumor tissue feature region after the influence of the ambient light is eliminated; the model training module is used for carrying out preliminary training on the selected characteristic wave bands through a semi-supervised clustering algorithm, and then training convolutional neural networks of different dimensions through transfer learning; a tumor tissue recognition model is obtained; and the tumor identification module is configured