Weld seam tracking is the key technology for achieving automated welding. In this paper, a seam tracking method based on a line array camera and a feature recognition algorithm is constructed to provide gapless seam recognition under laser welding. The workpiece surface image is captured by a line array camera and denoised by mean filtering. By analyzing the gray value of the image, the change in the gray value of the image, the gray gradient of the image, and the width of the weld area, the weld track is identified, and the identification and tracking of the gapless weld are realized. Welding experiments are conducted, and the results show that the error of the weld identification method based on a line array camera is less than 0.05 mm. Weld seam tracking experiments on welds with different a gap are conducted, and the results show that the proposed method has a good recognition effect for welds with a gap less than 0.3 mm.