Wall Cracks Detection in Aerial Images Using Improved Mask R-CNN

The present paper proposes a detection method for building exterior wall cracks since manual detection methods have high risk and low efficiency.The proposed method is based on Unmanned Aerial Vehicle(UAV)and computer vision technology.First,a crack dataset of 1920 images was established using UAV to collect the images of a residential building exterior wall under different lighting conditions.Second,the average crack detection precisions of different methods including the Single Shot MultiBox Detector,You Only Look Once v3,You Only Look Once v4,Faster Regional Convolutional Neural Network(R-CNN)and Mask R-CNN methods were compared.Then,the Mask R-CNN method with the best performance and average precision of 0.34 was selected.Finally,based on the characteristics of cracks,the utilization ratio of Mask R-CNN to the underlying features was improved so that the average precision of 0.9 was achieved.It was found that the positioning accuracy and mask coverage rate of the proposed Mask R-CNN method are greatly improved.Also,it will be shown that using UAV is safer than manual detection because manual parameter setting is not required.In addition,the proposed detection method is expected to greatly reduce the cost and risk of manual detection of building exterior wall cracks and realize the efficient identification and accurate labeling of building exterior wall cracks.

Safety Helmet Wearing Detection in Aerial Images Using Improved YOLOv4

In construction,it is important to check whether workers wear safety helmets in real time.We proposed using an unmanned aerial vehicle(UAV)to monitor construction workers in real time.As the small target of aerial photography poses challenges to safety-helmet-wearing detection,we proposed an improved YOLOv4 model to detect the helmet-wearing condition in aerial photography:(1)By increasing the dimension of the effective feature layer of the backbone network,the model’s receptive field is reduced,and the utilization rate of fine-grained features is improved.(2)By introducing the cross stage partial(CSP)structure into path aggregation network(PANet),the calculation amount of themodel is reduced,and the aggregation efficiency of effective features at different scales is improved.(3)The complexity of the YOLOv4 model is reduced by introducing group convolution and the pruning PANet multi-scale detection mode for de-redundancy.Experimental results show that the improved YOLOv4 model achieved the highest performance in the UAV helmet detection task,that the mean average precision(mAP)increased from83.67%of the original YOLOv4 model to 91.03%,and that the parameter amount of the model is reduced by 24.7%.The results prove that the improved YOLOv4 model can effectively respond to the requirements of real-time detection of helmet wearing by UAV aerial photography.