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 a significant role in the history of modern Chinese architecture. However, the recognition and classification of the qilou mainly rely on manual inspection...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2024-03-01
|
| Series: | Journal of Asian Architecture and Building Engineering |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1080/13467581.2023.2238038 |
| _version_ | 1827324157495345152 |
|---|---|
| author | Ming Ho Li Yi Yu Hongni Wei Ting On Chan |
| author_facet | Ming Ho Li Yi Yu Hongni Wei Ting On Chan |
| author_sort | Ming Ho Li |
| collection | DOAJ |
| description | Qilou (arcade building) is a particular type of Chinese historical architecture combined with western and eastern building elements, which plays a significant role in the history of modern Chinese architecture. However, the recognition and classification of the qilou mainly rely on manual inspection, suppressing the cultural dissemination and protection of qilou relics. In this paper, we present a new framework that adopts multiple image processing algorithms and a deep learning network to automate qilou classification. First, image dataset of the qilou is enhanced based on the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm. Then, an improved Faster R-CNN with ResNet50 (Faster R-CNN-R) is deployed for qilou image recognition. A total of 760 images captured in Guangzhou were used for training, validation, and accuracy check of the proposed framework and several contrastive networks under the same conditions. Compared to other networks, the proposed framework works better than Faster R-CNN with VGG16 (Faster R-CNN-V) and FCOS. The accuracy of the proposed framework embedded with the Faster R-CNN-R, Faster R-CNN-V, and FCOS are 80.12%, 65.17%, and 66.35%, respectively. Based on digital images captured under different lighting conditions, the proposed framework can be used to classify nine different types of qilous, with high robustness. |
| first_indexed | 2024-03-12T17:49:43Z |
| format | Article |
| id | doaj.art-ae0562e65a5d4712be580d2613d30d5f |
| institution | Directory Open Access Journal |
| issn | 1347-2852 |
| language | English |
| last_indexed | 2024-04-25T02:07:34Z |
| publishDate | 2024-03-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Journal of Asian Architecture and Building Engineering |
| spelling | doaj.art-ae0562e65a5d4712be580d2613d30d5f2024-03-07T14:28:18ZengTaylor & Francis GroupJournal of Asian Architecture and Building Engineering1347-28522024-03-0123259561210.1080/13467581.2023.22380382238038Classification of the qilou (arcade building) using a robust image processing framework based on the Faster R-CNN with ResNet50Ming Ho Li0Yi Yu1Hongni Wei2Ting On Chan3Sun Yat-sen UniversityEast China Normal UniversityGuangdong University of Foreign StudiesSun Yat-sen UniversityQilou (arcade building) is a particular type of Chinese historical architecture combined with western and eastern building elements, which plays a significant role in the history of modern Chinese architecture. However, the recognition and classification of the qilou mainly rely on manual inspection, suppressing the cultural dissemination and protection of qilou relics. In this paper, we present a new framework that adopts multiple image processing algorithms and a deep learning network to automate qilou classification. First, image dataset of the qilou is enhanced based on the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm. Then, an improved Faster R-CNN with ResNet50 (Faster R-CNN-R) is deployed for qilou image recognition. A total of 760 images captured in Guangzhou were used for training, validation, and accuracy check of the proposed framework and several contrastive networks under the same conditions. Compared to other networks, the proposed framework works better than Faster R-CNN with VGG16 (Faster R-CNN-V) and FCOS. The accuracy of the proposed framework embedded with the Faster R-CNN-R, Faster R-CNN-V, and FCOS are 80.12%, 65.17%, and 66.35%, respectively. Based on digital images captured under different lighting conditions, the proposed framework can be used to classify nine different types of qilous, with high robustness.http://dx.doi.org/10.1080/13467581.2023.2238038qilouobject detectionfaster r-cnnclaheclassification |
| spellingShingle | Ming Ho Li Yi Yu Hongni Wei Ting On Chan Classification of the qilou (arcade building) using a robust image processing framework based on the Faster R-CNN with ResNet50 Journal of Asian Architecture and Building Engineering qilou object detection faster r-cnn clahe classification |
| title | Classification of the qilou (arcade building) using a robust image processing framework based on the Faster R-CNN with ResNet50 |
| title_full | Classification of the qilou (arcade building) using a robust image processing framework based on the Faster R-CNN with ResNet50 |
| title_fullStr | Classification of the qilou (arcade building) using a robust image processing framework based on the Faster R-CNN with ResNet50 |
| title_full_unstemmed | Classification of the qilou (arcade building) using a robust image processing framework based on the Faster R-CNN with ResNet50 |
| title_short | Classification of the qilou (arcade building) using a robust image processing framework based on the Faster R-CNN with ResNet50 |
| title_sort | classification of the qilou arcade building using a robust image processing framework based on the faster r cnn with resnet50 |
| topic | qilou object detection faster r-cnn clahe classification |
| url | http://dx.doi.org/10.1080/13467581.2023.2238038 |
| work_keys_str_mv | AT mingholi classificationoftheqilouarcadebuildingusingarobustimageprocessingframeworkbasedonthefasterrcnnwithresnet50 AT yiyu classificationoftheqilouarcadebuildingusingarobustimageprocessingframeworkbasedonthefasterrcnnwithresnet50 AT hongniwei classificationoftheqilouarcadebuildingusingarobustimageprocessingframeworkbasedonthefasterrcnnwithresnet50 AT tingonchan classificationoftheqilouarcadebuildingusingarobustimageprocessingframeworkbasedonthefasterrcnnwithresnet50 |