A new efficient multiobject detection and size calculation for blended tobacco shred using an improved YOLOv7 network and LWC algorithm

Abstract

Abstract Detection of the four tobacco shred varieties, including tobacco silk, cut stem, expended tobacco silk, and reconstituted tobacco shred, and the subsequent calculation of the tobacco shred component ratio and unbroken tobacco shred rate are the primary tasks in cigarette inspection lines. The accuracy, speed and recognizable complexity of tobacco shred images affect the feasibility of practical applications directly in the inspection line field. In cigarette quality inspection lines, there are bound to be a large number of single tobacco shreds and a certain amount of overlapped tobacco shreds at the same time, and it is especially critical to identify both single and overlapped tobacco shreds at once, that is, fast blended tobacco shred detection based on multiple targets. However, it is difficult to classify tiny single tobacco shreds with complex morphological characteristics, not to mention classifying and locating tobacco shreds with 24 types of overlap alone, which poses significant difficulties for machine vision-based blended tobacco shred multiobject detection and unbroken tobacco shred rate calculation tasks. This study focuses on the two challenges of identifying blended tobacco shreds with single tobacco shreds and overlapped tobacco simultaneously in the field application and calculating the unbroken tobacco shred rate. In this paper, a new multiobject detection model is developed for blended tobacco shred images based on an improved YOLOv7-tiny. YOLOv7-tiny is used as the mainframe of the multiobject detection network. The lightweight ResNet19 is used as the model backbone. The original SPPCSPC and coupled detection head are replaced with a new spatial pyramid SPPFCSPC and a decoupled joint detection head, respectively. An algorithm for the two-dimensional size calculation of the blended tobacco shred (LWC) is also proposed, which is applied to blended tobacco shred object detection images to obtain independent tobacco shred objects and calculate the unbroken tobacco shred rate. The experimental results showed that the final detection precision, mAP@.5, mAP@.5:.95, and testing time were 0.883, 0.932, 0.795, and 4.12 ms, respectively. The average length and width detection accuracies of blended tobacco shred samples were -1.7% and 13.2%, respectively. It achieved high multiobject detection accuracy and 2D dimensional size calculation accuracy, which also conformed to the manual inspection process in the field. This study provides a new efficient implementation method for multiobject detection and size calculation of blended tobacco shreds in the field of cigarette quality inspection lines and a new approach for other similar blended image multiobject detection tasks.