During the development process of parachutes, a large amount of video image data is generated. By analyzing and processing the video data, many useful parachute parameters can be obtained, enriching the evaluation means of parachute performance. Therefore, the analysis of parachute video images is being paid more and more attention by researchers. However, in the parameter identification of parachute video images, the segmentation of canopy area is mainly based on traditional digital image processing algorithms. This method requires frequent adjustment of algorithm parameters to achieve canopy precise segmentation for different types of parachute images, not only leading to low segmentation efficiency, but also insufficient universality. Therefore, this paper introduces YOLOv8 for canopy segmentation, improving its universality and parachute parameter identification efficiency through YOLOv8's automatic and efficient segmentation capabilities. By comparing the canopy tilt angle, canopy projected perimeter, and canopy projected area, obtained by parameter identification based on traditional digital image processing algorithms and the proposed method, the results show that the proposed method effectively improves the efficiency of parachute parameter identification while maintaining the identification accuracy of parachute parameters.