The work aims to recognize and locate the chip by means of the chip recognition and location algorithm based on the improved Hu moments and rectangle fitting, for the purpose of realizing the visual recognition and location of the automatic SIM placement machine with high quality and accuracy. Firstly, the image was pre-processed, and the contour shape of the image was extracted based on the Ostu segmentation. Then, the chip was recognized and detected with the improved Hu moments combined with the relative moments and the eccentricity characteristics. Finally, the external contour was extracted on the basis of the recognition and the chip was located through the rectangle fitting of the external contour. The experimental results showed that, for chip recognition with an improved Hu moment, the accuracy was at least up to 98%, significantly improving the ability of chip recognition. When the chip was located and the rectangle fitting positioning accuracy was within 0.5 pixels, the position precision was high. The method can basically meet the requirements of high quality and high precision for the visual system of the SIM placement machine during chip placement.