The invention provides a cashmere and wool fiber identification method based on local characteristics and a word bag model, comprising the following steps: firstly, carrying out sample preparation andimage acquisition, and then preprocessing the image; Extracting local features from the fiber image, wherein the local features comprise image feature point detection, key point positioning, key point direction determination and key point description; Constructing a visual dictionary and image description, including generation of visual words and the visual dictionary, image description and spatial pyramid matching; And finally, generating a classification recognizer for performing binary classification recognition on the cashmere and wool fiber micrographs. The method provided by the invention overcomes the defects in the prior art, and can automatically extract and identify the morphological characteristics of the fiber surface, thereby objectively, accurately and quickly identifying the cashmere and wool fibers, and the method is high in identification accuracy and easy to operate.