The defect of the train wheel tread is a threat to its safe driving, and the defect detection of the tread is an important work. The extraction of defect area is a crucial link. In this paper, we propose a segmentation algorithm of tread defect area based on attention mechanism, which realizes the more accurate segmentation of tread defect area.This algorithm uses U-net as the backbone network, firstly, introduces the Lovasz-Softmax loss, secondly, CBAM is introduced between the encoder and decoder. Get the attention feature map information in the channel and space dimensions, and then multiply the two feature map information with the original input feature map to make adaptive feature correction to obtain a more accurate feature map and improve the accuracy of the segmentation algorithm.Validated on the dataset of train wheel tread, and the experimental results show that the algorithm PA is 99.54% and mIoU is 98.27%, which improves by 0.83% and 0.73% compared with Unet algorithm, which verifies the effectiveness of the algorithm.
End-to-end(E2E) scene text recognition (the joint detection and recognition of natural text images) is developing rapidly, and the joint optimization strategy for image sequence alignment has become a research hotspot. Those existing methods are either difficult to train or costly for character annotations. In this paper, a novel end-to-end scene text recognition framework is proposed, Based on the Swin-Transformer (Swin-T) FPN backbone network, the model adopts the instance segmentation method to obtain the text mask and binarizes it to directly locate its polygon boundaries. Meanwhile, to solve the problem of low fitting efficiency of the text sequence recognition module, we designed a self-monitoring Mask-Supervised Attention (MSA) mechanism to accelerate the fitting speed and fitting accuracy of the recognition module, thereby improving the joint performance for E2E text recognition. The results show that in the E2E text recognition task, the F-measure performance of the proposed model achieves not only 2.3%, 2.7% and 11.9% improvement on ICDAR 2015 on strong, weak and generic lexicons, but also 4.8%, 9.9% improvement on Total-text on full and none lexicons compared with other typical models.
Three dimensional (3D) shape reconstruction based on structured light technique is one of the most crucial and attractive techniques in the field of optical metrology and measurements due to the nature of non-contact and high-precision. Acquiring high-quality 3D shape data of objects with complex surface is an issue that is difficult to solve by single-frequency method. However, 3D shape data of objects with complex surface can be obtained only at a limited accuracy by classical multi-frequency approach. In this paper, we propose a new robust deep learning shape reconstruction (DLSR) method based on the structured light technique, where we accurately extract shape information of objects with complex surface from three fringe patterns with different frequencies. In the proposed DLSR method, the input of the network is three deformed fringe patterns, and the output is the corresponding 3D shape data. Compared with traditional approach, the DLSR method is pretty simple without using any geometric information and complicated triangulation computation. The experimental results demonstrate that the proposed DLSR method can effectively achieve robust, high-precision 3D shape reconstruction for objects with complex surface.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.