Salient object detection aims to identify the most prominent objects within an image. With the advent of fully convolutional networks (FCNs), deep learning-based saliency detection models have increasingly leveraged FCNs for pixel-level saliency prediction. However, many existing algorithms face challenges in accurately delineating target boundaries, primarily due to insufficient utilization of edge information. To address this issue, we propose a novel approach to improve the boundary accuracy of salient target detection by integrating salient target and edge information. Our approach comprises two key components: a Self-attentive Group Pixel Fusion module (SGPFM) and a Bidirectional Feature Fusion module (BFF). The SGPFM extracts salient edge features from the lower layers of ResNet50 and salient target features from the higher layers. These features are then optimized using a self-attentive mechanism. The BFF module progressively fuses the salient target and edge features, optimizing them based on their logical relationships and enhancing the complementarities among the features. By combining detailed edge information and positional target information, our method significantly enhances the detection accuracy of target boundaries. Experimental results demonstrate that the proposed model outperforms the latest existing methods across four benchmark datasets, providing accurate and detail-rich salient target predictions. This advancement marks a significant contribution to the development of the field.

As the primary carrier for oil and gas transportation, pipelines are critical for the entire industry. Pipelines are continuously subjected to corrosion and abrasion in the oil and gas delivery process, leading to gradual wall thickness reduction, shortened service life, and deteriorated operational safety. Ultrasonic testing has been widely adopted for monitoring pipeline wall thickness. Conventional ultrasonic transducers possess rigid configurations, which hinder large-area inspection and exhibit poor adaptability to complex curved components. In contrast, flexible ultrasonic sensors show prominent advantages, with their small size, light weight, and excellent conformal contact with curved surfaces. Flexible piezoelectric thin-film sensors have been used in a wide range of fields. As one of the most representative piezoelectric polymers, poly(vinylidene fluoride–trifluoroethylene) (P(VDF-TrFE)) combines favorable piezoelectric coefficients and intrinsic flexibility, making it popular. Some research groups have investigated the influences of modified filler particles, doping ratios, and fabrication process optimization on the performance of P(VDF-TrFE)-based piezoelectric composites, while others have concentrated on the practical applications of existing flexible piezoelectric sensors. This study emphasizes a rapid customized fabrication strategy for flexible sensors instead of single-specification standardized probes; hence, it does not share the same comparison benchmark as conventional fixed-dimension sensors. Systematic research on flexible piezoelectric thin-film sensors is presented, including piezoelectric material modification, substrate design, laminated structural design, fabrication workflows, establishment of the testing platform, and the development of matched circuit systems. The material preparation and manufacturing processes are optimized, and a scalable technical route for fabricating flexible piezoelectric sensors is proposed. Using this route, flexible piezoelectric thin-film sensors can be rapidly tailored for different application scenarios to satisfy diverse engineering demands. Multiple experiments were conducted on pipeline samples with varying wall thicknesses and curvatures. The results verify that the sensor reaches a measurement precision of 0.01 mm, meeting the demands of high-precision pipeline structural health monitoring.