Skip to main content
ARS Home » Plains Area » Lubbock, Texas » Cropping Systems Research Laboratory » Cotton Production and Processing Research » Research » Publications at this Location » Publication #389228
Research Project: Enhancing the Profitability and Sustainability of Upland Cotton, Cottonseed, and Agricultural Byproducts through Improvements in Pre-Ginning, Ginning, and Post-Ginning Processes

Location: Cotton Production and Processing Research

Title: A CNN-based approach to detect cover damage of round cotton modules

Author
item IQBAL, MD - Texas A&M University
item HARDIN, ROBERT - Texas A&M University
item WANG, TIANYI - Texas A&M University
item WARD, JASON - North Carolina State University
item Wanjura, John

Submitted to: National Cotton Council Beltwide Cotton Conference
Publication Type: Abstract Only
Publication Acceptance Date: 10/21/2021
Publication Date: 10/21/2021
Citation: Iqbal, M.Z., Hardin, R.G., Wang, T., Ward, J.K., Wanjura, J.D. 2021. A CNN-based approach to detect cover damage of round cotton modules. National Cotton Council Beltwide Cotton Conference.

Interpretive Summary:

Technical Abstract: Round cotton modules covered with engineered plastic film are increasingly popular because they largely automate the cotton harvest. During the handling of the round cotton modules, it was indicated that the cover damage might occur frequently which is the major source of plastic contamination in the processed cotton fiber. Plastic contamination costs millions of dollars of premiums every year to the US cotton industry, whereas this industry was one of the cleanest cotton providers to the international market. To uphold the reputation, the US cotton industries are putting high priority for the removal of plastic contamination from cotton. To produce contamination-free fiber, it is important to determine detailed information about the plastic cover damage of the cotton modules. Farmers and ginners might want to identify damaged covers to be able to repair or handle carefully. An automatic cover damage identification system might be able to resolve this issue. Therefore, the objective of this research was to develop a convolutional neural network (CNN) based model to classify the cotton modules with damaged covers during the handling process. To collect images of the cotton modules during the handling process, a single-board computer-based system was developed and installed on a loader. This system was able to collect images of the cotton modules from different directions during the handling process and stored them with a unique RFID number. A CNN model was developed by using some training and testing images of cotton modules having damaged and undamaged covers. This model was able to provide information about the status of each module from the images collected by the system. The gathered information from this study might be helpful to produce contamination-free high-quality cotton by improving the handling, and ginning process consequently.