Component surface defect detection based on image segmentation method

Renwei, Lu; Dong, Yin

doi:10.1109/ccdc.2016.7531906

Cited by 2 publications

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the improvement of computer computing power and breakthroughs in neural network technology, object detection of illegal buildings in ancient cities has been widely applied in the field of deep learning. Zhang Tong [1] from Wuhan University according to remote sensing images proposed automatic building detection algorithm, applying deep learning to the field of unauthorized building identification; Dong Renwei [2] et al proposed a Faster RCNN-based aerial camera images of UAVs and DCGAN for small-sample enhanced unauthorized building target detection, providing a theoretical and application basis for fast and accurate urban unauthorized building detection; Zheheng Liang, Peng Deng [3] team implement a neural network-based approach unauthorized building detection algorithm for UAV images, with an accuracy and recall rate of 71% and 88% respectively. Although deep learning technology has achieved good results in detecting illegal building targets, there is still much room for improvement in both the accuracy and speed of its detection.…”

Section: Introductionmentioning

confidence: 99%

YOLO-UB Based Detection and Identification of Illegal Structures in the Ancient City

Ci,

Li,

Sun

et al. 2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

To address the problem that current target detection algorithms do not work well in the task of detecting illegal building targets in the ancient city, a YOLO-UB-based algorithm for illegal building detection and recognition in the ancient city is proposed based on remote sensing image data taken by UAV. The algorithm incorporates the Coordinate attention (CA) attention mechanism to improve network robustness enhance the model’s detection ability for illegal building targets, and make target localization more accurate, and introduces the Swin Transformer V2 structure to use its own self-attention mechanism to deeply mine the target features, which enhances the global information capture capability and enables the network to better integrate multi-scale features. The algorithm is trained iteratively on a custom dataset and compared with other models. The results show that this algorithm achieves an average accuracy (mAP) of 96.8% in detecting illegal building targets, Compared to the algorithm with YOLOv7, the accuracy is improved by 3.1%, and the algorithm has better featureextraction, robustness and generalization than other target detection models.

show abstract

Section: Introductionmentioning

confidence: 99%