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 385 f/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.

Retinopathy is an important ophthalmic disease that causes blindness in the elderly population. As the global elderly demographic expands, the importance of the efficient ophthalmic healthcare system for pre-diagnosis cannot be overstated. Optical coherence tomography (OCT) is considered the gold standard for ophthalmic treatment and diagnosis. OCT technologies and equipment continue to develop towards the intelligence and convenience for requirements of rapid diagnosis in the remote and poverty-stricken areas. Here, we proposed an improved MobilenetV2 lightweight model for retinopathy classification (MobilenetV2-RC), which incorporates spatial and channel reconstruction convolution and the improved convolutional block attention module attention mechanism into the framework. Not only can it effectively limit feature redundancy to reduce model parameters, but also enhance the ability of feature representation to improve classification accuracy. The parameters of the proposed model are only 1.96 M with an overall accuracy of 98.96%, which is higher 3.32% than the original MobilenetV2. Compared with ResNet18, InceptionV3, and VGG16_BN, the overall accuracy is increased by 4.6%, 6.3%, and 3.9%, respectively. The test results of UCSD and Duke open-source datasets are more remarkable. Experimental results show that our proposed algorithm has strong reliability and generalization for the accurate classification of retinopathy, and a greater application prospect in the intelligent diagnosis of ophthalmology and mobile detection terminals.

To improve the calibration efficiency and accuracy of line structured light sensor, a plane checkerboard and concentric circle complementary line structured light calibration system was designed integrating with self-backlighting and pose-adjustable functions. Based on the geometric relationship between the real projection position of the concentric circle center and the center of projection ellipse, the nonlinear optimized compensation model for eccentric error was established, which precisely determined the eccentric error compensation position of the circle center under the perspective projection. Compared with the traditional calibration method, the proposed method reduced the re-projection error by 84.7% and effectively solved the problem of high-precision calibration for eccentric error compensation of circular markers. By combining the plane passing through the optical center and the center line of the light bar with the target plane in the camera coordinate system, the coordinate information of the spatial intersection lines was obtained for many times to increase the feature points, and the problem of lower calibration precision of plane fitting due to the fewer feature points was solved by using the least square method to fit the optic plane equation. In complex environments, the average error of the outer diameter of the large-size grinding wheel is 0.005 1 mm measured by repeated experiments, and the results show that the designed calibration system has certain precision, simplicity, and practicability.