Monitoring biophysical signals such as body or organ movements and other physical phenomena is necessary for patient rehabilitation. However, stretchable flexible pressure sensors with high sensitivity and a broad range that can meet these requirements are still lacking. Herein, we successfully monitored various vital biophysical features and implemented in-sensor dynamic deep learning for knee rehabilitation using an ultrabroad linear range and high-sensitivity stretchable iontronic pressure sensor (SIPS). We optimized the topological structure and material composition of the electrode to build a fully stretching on-skin sensor. The high sensitivity (12.43 kPa−1), ultrabroad linear sensing range (1 MPa), high pressure resolution (6.4 Pa), long-term durability (no decay after 12000 cycles), and excellent stretchability (up to 20%) allow the sensor to maintain operating stability, even in emergency cases with a high sudden impact force (near 1 MPa) applied to the sensor. As a practical demonstration, the SIPS can positively track biophysical signals such as pulse waves, muscle movements, and plantar pressure. Importantly, with the help of a neuro-inspired fully convolutional network algorithm, the SIPS can accurately predict knee joint postures for better rehabilitation after orthopedic surgery. Our SIPS has potential as a promising candidate for wearable electronics and artificial intelligent medical engineering owing to its unique high signal-to-noise ratio and ultrabroad linear range.
Counting the number of wheat ears in images under natural light is an important way to evaluate the crop yield, thus, it is of great significance to modern intelligent agriculture. However, the distribution of wheat ears is dense, so the occlusion and overlap problem appears in almost every wheat image. It is difficult for traditional image processing methods to solve occlusion problem due to the deficiency of high-level semantic features, while existing deep learning based counting methods did not solve the occlusion efficiently. This article proposes an improved EfficientDet-D0 object detection model for wheat ear counting, and focuses on solving occlusion. First, the transfer learning method is employed in the pre-training of the model backbone network to extract the high-level semantic features of wheat ears. Secondly, an image augmentation method Random-Cutout is proposed, in which some rectangles are selected and erased according to the number and size of the wheat ears in the images to simulate occlusion in real wheat images. Finally, convolutional block attention module (CBAM) is adopted into the EfficientDet-D0 model after the backbone, which makes the model refine the features, pay more attention to the wheat ears and suppress other useless background information. Extensive experiments are done by feeding the features to detection layer, showing that the counting accuracy of the improved EfficientDet-D0 model reaches 94%, which is about 2% higher than the original model, and false detection rate is 5.8%, which is the lowest among comparative methods.
Abstract:In order to automatically evaluate the welding quality during high-power disk laser welding, a real-time monitoring system was developed. The images of laser-induced metal vapor during welding were captured and fifteen features were extracted. A feature selection method based on a sequential forward floating selection algorithm was employed to identify the optimal feature subset, and a support vector machine (SVM) classifier was built to recognize the welding quality. The experiment results demonstrated that this method had satisfactory performance, and could be applied in laser welding monitoring applications.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.