Classification of the qilou (arcade building) using a robust image processing framework based on the Faster R-CNN with ResNet50

Qilou (arcade building) is a particular type of Chinese historical architecture combined with western and eastern building elements, which plays significant role in the history modern architecture. However, recognition classification qilou mainly rely on manual inspection, suppressing cultural dissemination protection relics. In this paper, we present new framework that adopts multiple image processing algorithms deep learning network to automate classification. First, dataset enhanced based Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm. Then, an improved Faster R-CNN ResNet50 (Faster R-CNN-R) deployed for recognition.Seven hundred sixty images captured Guangzhou, were used training, validation, accuracy check proposed several contrastive networks under same conditions. Compared other networks, works better than VGG16 R-CNN-V) FCOS. The embedded R-CNN-R, R-CNN-V, FCOS are 80.12%, 65.17% 66.35%, respectively. Based digital different lighting conditions, can be classifly nine qilou, high robustness.