In view of the characteristics of diverse shapes, frequent updates and small batches of products in small and micro ceramic enterprises, the existing robotic arm glazing system has problems such as high cost, complex maintenance, and repeated programming for different types of equipment. This paper proposes a dynamic template matching glazing algorithm based on FPGA (FDTM-GS). By collecting the background image and the reference image at one time, leveraging dynamic template matching between a 9×9 background template and sliding 4×4 real-time submatrices to quantify positional offsets, the algorithm realized the rapid positioning of the bisque body. Gaussian filtering and dynamic threshold segmentation are used to eliminate the environmental noise, and the glaze path is generated according to the pixel-physical size mapping relationship. The system uses FPGA hardware to implement image processing algorithms, which reduces resource consumption and supports off-line operation without additional programming for new devices. The experimental results show that the standard deviation of coating thickness uniformity is 0.025mm (n=42), and the comprehensive cost is only 1/7-1/10 of the manipulator scheme, which meets the requirements of daily ceramic technology.