It is important to detect and classify foreign fibers in cotton, especially white and transparent foreign fibers, to produce subsequent yarn and textile quality. There are some problems in the actual cotton foreign fiber removing process, such as some foreign fibers missing inspection, low recognition accuracy of small foreign fibers, and low detection speed. A polarization imaging device of cotton foreign fiber was constructed based on the difference in optical properties and polarization characteristics between cotton fibers. An object detection and classification algorithm based on an improved YOLOv5 was proposed to achieve small foreign fiber recognition and classification. The methods were as follows: (1) The lightweight network Shufflenetv2 with the Hard-Swish activation function was used as the backbone feature extraction network to improve the detection speed and reduce the model volume. (2) The PANet network connection of YOLOv5 was modified to obtain a fine-grained feature map to improve the detection accuracy for small targets. (3) A CA attention module was added to the YOLOv5 network to increase the weight of the useful features while suppressing the weight of invalid features to improve the detection accuracy of foreign fiber targets. Moreover, we conducted ablation experiments on the improved strategy. The model volume, mAP@0.5, mAP@0.5:0.95, and FPS of the improved YOLOv5 were up to 0.75 MB, 96.9%, 59.9%, and 385f/s, respectively, compared to YOLOv5, and the improved YOLOv5 increased by 1.03%, 7.13%, and 126.47%, respectively, which proves that the method can be applied to the vision system of an actual production line for cotton foreign fiber detection.

Lint percentage of seed cotton is one of the important bases for pricing in the trading segment. Unfortunately, the conventional methods of lint percentage are manually operated, which relies on the abundant experience of experts, and restrained by personal, physical and environmental factors. Up to date, the calculation of the lint percentage of seed cotton has not fully automated. In this paper, we proposed a non-destructive detection method for automatically obtaining lint percentage of seed cotton based on optical penetration imaging and machine vision, for the first time to our knowledge. The cotton seed image was obtained by the penetration imaging setup with a LED white backlight source. To accurately identify the number of cotton seeds, the image features of the cotton seed was studied and three key features was been found, which are the circumference, area, and greyscale value, respectively. A calculation system based on the three key features was presented to process the images and then automatically calculate the lint percentage of seed cotton. The first step of the system is to segment the original image using adaptive thresholding followed by morphological operations. Afterwards, the number of cotton seed was obtained by the three key features of the cotton seed. Then, the lint percentage was achieved by a professional industry formula. The suggested lint percentage detection methods were verified by the experiments with two seed cotton varieties samples of H219 and ZHM19. The experimental results indicated that the detection average accuracy of the developed system for seed cotton varieties H219 and ZHM19 were 96.33% and 95.40%

No abstract

No abstract