Method of tobacco volume measurement based on image segmentation and feature extraction and its application

In this paper, a method of tobacco volume measurement based on image segmentation and feature extraction is proposed. The tobacco image is collected by a high-resolution industrial camera, and the image quality is optimized by preprocessing techniques such as median filtering, histogram equalization and geometric correction. Image segmentation adopts an algorithm based on threshold segmentation and morphological processing, which can effectively separate the cut tobacco from the background and keep the complete contour of the cut tobacco. Feature extraction uses Canny edge detection algorithm to obtain tobacco contour information, and calculates the area by counting the number of pixels, and measures the longest distance in the main direction as the length. Combined with camera calibration parameters, tobacco layers and average thickness, a simplified cylinder volume calculation model was used to estimate tobacco volume. The experimental results show that this method can accurately measure the volume of different brands and types of cut tobacco, which is highly consistent with the actual volume, and its feasibility and reliability are verified. This method has the advantages of non-contact, high efficiency and easy realization, and can be widely used in tobacco industry to measure the volume of cut tobacco, providing technical support for improving production efficiency and product quality.