Nanocomposites offer a unique material choice for enhancing sensor sensitivity and stability in electrochemical measurements. Nowadays, the feature extraction and image processing of nanocomposites in electrochemical determination have serious problems of error rate and low efficiency, which have a great influence on the experimental process. K-means algorithm is mostly used for image feature extraction. In this paper, K-means algorithm is used to extract and process images of nanocomposites in electrochemical determination effectively, and conclusions are drawn through experiments. The algorithm greatly reduces the error rate of feature extraction and image processing, and the accuracy rate is increased to 99.4%.