AUTHOR=Wang Li , Jia Kunming , Fu Yongmin , Xu Xiaoguang , Fan Lei , Wang Qiao , Zhu Wenkui , Niu Qunfeng 

TITLE=Overlapped tobacco shred image segmentation and area computation using an improved Mask RCNN network and COT algorithm

JOURNAL=Frontiers in Plant Science

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2023.1108560

DOI=10.3389/fpls.2023.1108560

ISSN=1664-462X

ABSTRACT=Classification of the four tobacco shred varieties, including tobacco silk, cut stem, expended tobacco silk, and reconstituted tobacco shred, and the subsequent determination of tobacco shred components are the primary tasks for calculating the tobacco shred blending ratio. The identification accuracy and subsequent component area calculation error directly affect the composition determination and quality of the tobacco shred. However, tiny tobacco shreds have complex physical and morphological characteristics, especially the substantial similarity between expended tobacco silk and tobacco silk. It complicates their classification. There must be a certain amount of overlap and stacking in the distribution of tobacco shreds on the actual tobacco quality inspection line. There are 24 types of overlap alone, not to mention the stacking phenomenon. The self-winding of its varieties does not make them easier to distinguish from the overlapped types, posing significant difficulties for machine vision-based tobacco shred classification and component area calculation tasks. This study focuses on two significant challenges associated with identifying various types of overlapping tobacco shreds and acquiring overlapping regions to calculate overlapping areas. It develops a new segmentation model for tobacco shred images based on an improved Mask RCNN network. Mask RCNN is used as the segmentation network’s mainframe. Convolutional network and FPN in Backbone are replaced with Densenet121 and U-FPN, respectively. The size and aspect_ratios of Anchors in RPN are optimized. An algorithm for area calculation of overlapped tobacco shred region (COT) is also proposed, which is applied to overlapped tobacco shred mask images to obtain overlapped regions and calculate the overlapped area. The experimental results showed that the final segmentation accuracy and recall rates are 89.1% and 73.2%, respectively. The average area detection rate of 24 overlapped tobacco shred samples increases from 81.2% to 90%, achieving high segmentation accuracy and overlapped area calculation accuracy. This study provides a new implementation method for the type identification and component area calculation of overlapped tobacco shreds and a new approach for other similar overlapped image segmentation tasks.