DMR-RTDETR: A Multiscale and Context-Aware Measurement Framework for Quantitative Characterization of Photovoltaic Module Thermal Anomalies With Uncertainty Analysis

The image data acquired through Unmanned Aerial Vehicle (UAV)-based infrared imaging typically features complex backgrounds. Moreover, photovoltaic (PV) faults appear as small, low-contrast anomalies, making them difficult to detect accurately using conventional methods. In this paper, we design a Dual-path Multi-scale and Reparametrized Real-Time Detection Transformer (DMR-RTDETR) network, which mainly consists of Multi-Path Cross-Scale Attention Feature Fusion (MPCAFF) module, RepshuffleRoss module, and optimized prediction head. First, the MPCAFF module recalibrates and fuses global and local feature maps in parallel to enhance the detection of small faults. Second, the RePshuffleRoss Reparameterization Feature Shuffling module uses multi-path convolutions and channel shuffling to boost feature diversity, while reparameterization during inference reduces computational complexity. Third, an optimized RepConv-based detection head is designed to streamline the decoder and enhance accuracy by selectively integrating 40×40 and 80×80 scale feature maps. In addition, we also propose the Photovoltaic Fault Detection (PVFD) dataset, which is the first target detection database covering multiple PV fault types. Experimental results show that the DMR-RTDETR model has a mAP of 0.871 and an APs of 0.448 on the PVFD dataset, confirming its effectiveness in detecting small target faults, typically defined as occupying less than 32 × 32 pixels. Finally, through comparative experiments on the public VisDrone2019 benchmark dataset, the generalization ability of the proposed model is further verified.