Fabric quality plays a crucial role in modern textile industry processes. How to detect fabric defects quickly and effectively has become the main research goal of researchers. The You Only Look Once (YOLO) series of networks have maintained a dominant position in the field of target detection. However, detecting small-scale objects, such as tiny targets in fabric defects, is still a very challenging task for the YOLOv4 network. To address this challenge, this paper proposed an improved YOLOv4 target detection algorithm: using a combined data augmentation method to expand the dataset and improve the robustness of the algorithm, obtaining the anchors suitable for fabric defect detection by using the k-means algorithm to cluster the ground truth box of the dataset, adding a new prediction layer in yolo_head in order to have a better effect on tiny target detection, integrating a convolutional block attention module into the backbone feature extraction network, and innovatively replacing the CIOU loss function with the CEIOU loss function to achieve accurate classification and localization of defects. Experimental results show that compared with the original YOLOv4 algorithm, the detection accuracy of the improved YOLOv4 algorithm for tiny targets has been greatly increased, the AP value of tiny target detection has increased by 12%, and the overall mean average precision (mAP) has increased by 3%. The prediction results of the proposed algorithm can provide enterprises with more accurate defect positioning, reduce the defect rate of fabric products, and improve their economic effect.

Deep learning has been applied in various fields for its effective and accurate feature learning capabilities in recent years. Currently, information extracted from remote sensing images with the learning methods has become the most relevant research area for its developed precision. In terms of developing segmentation precision and reducing calculation power consumption, the improved deep learning methods have received more attention, and the improvement of semantic segmentation architectures has been a popular solution. This research presents a learning method named D-DenseNet with a new structure for road extraction. The methods for the improvement are divided into two stages: (1) alternate the consecutive dilated convolutions number in the structure of the network (2) the stem block is arranged as the initial block. So, dilated convolution can obtain more global context information through the whole network. Further, the D-DenseNet restructures D-LinkNet by taking DenseNet as its backbone instead of ResNet, which can expand the receptive field and accept more feature information. The D-DenseNet is effective because of its 119 M model size and 57.96% IoU on the processing test data and 99.3 M modes size and 66.26% on the public dataset, which achieved the research objective for reducing model size and developing segmentation precision—IoU. The experiment indicates that the D-Dense block and the stem block are effective for developing road extraction, and the appropriate number of convolution layers is also essential for model evaluation.

Phase measurement profilometry (PMP) is primarily employed to analyze the morphology of a functional surface with precision. Historically, one of the most complex and persistent challenges in PMP has been reducing errors stemming from inconsistent indicators at the edges of a surface. In response to this challenge, we propose an optimized error compensation methodology specifically designed to handle edge artefacts. This methodology introduces the Hilbert transform and object surface albedo as tools to detect the edges of the artefact region that need to be compensated. Moreover, we analyze the characteristics of the sinusoidal fringe waveform propagation direction and investigate the reconstruction results of the fringe vertical to the current directions to compensate for edge artefacts. The experimental results for various objects show that the optimized approach can compensate for edge artefacts by projecting in two directions and reducing the projection by half. The compensated root mean square error (RMSE) for planar objects can be reduced by over 45%.

A Multi Secret Image sharing scheme can share several secret images among certain participators securely. Boolean-based secret sharing schemes are one kind of secret sharing method with light-weighted computation compared to the previous complex algebraic-based methods, which can realize the sharing of multi secret images. However, the existing Boolean-based multi secret sharing schemes are mostly restricted to the particular case of (2, n) and (n, n), only few Boolean-based multi secret sharing schemes study the general access structure, and the shares are mostly meaningless. In this paper, a new Boolean-based multi secret sharing scheme with the general access structure is proposed. All the shares are meaningful, which can avoid attracting the attention of adversaries, and the secret images can be recovered in a lossless manner. The feasibility of the scheme is proven, the performance is validated by the experiments on the gray images, and the analysis of the comparison with other methods is also given out.