Multi-Class Parrot Image Classification Including Subspecies with Similar Appearance

Abstract

Owing to climate change and human indiscriminate development, the population of endangered species has been decreasing. To protect endangered species, many countries worldwide have adopted the CITES treaty to prevent the extinction of endangered plants and animals. Moreover, research has been conducted using diverse approaches, particularly deep learning-based animal and plant image recognition methods. In this paper, we propose an automated image classification method for 11 endangered parrot species included in CITES. The 11 species include subspecies that are very similar in appearance. Data images were collected from the Internet and built in cooperation with Seoul Grand Park Zoo to build an indigenous database. The dataset for deep learning training consisted of 70% training set, 15% validation set, and 15% test set. In addition, a data augmentation technique was applied to reduce the data collection limit and prevent overfitting. The performance of various backbone CNN architectures (i.e., VGGNet, ResNet, and DenseNet) were compared using the SSD model. The experiment derived the test set image performance for the training model, and the results show that the DenseNet18 had the best performance with an mAP of approximately 96.6% and an inference time of 0.38 s.

Image Classification of Amazon Parrots by Deep Learning: A Potentially Useful Tool for Wildlife Conservation

“…This is because ResNet increases the computing speed by skip connection [ 42 ]. These proportion relationships between the performance of the object detection model and the complexity of CNN architectures used as backbone networks have been reported in previous studies [ 37 , 39 ].…”

“…The images of 35 adult Amazon parrots were collected from the Internet ( accessed on 10 March 2022) because there was no standard dataset for these species. The image collection from the Internet has been used to establish a dataset for deep learning to obtain images of various individuals with diverse backgrounds when a standard dataset is not available [ 37 , 39 ]. For comprehensive image collection, the species and common names were used as keywords.…”

“…Moreover, SSD shows faster data processing speed and accuracy than YOLO because SSD performs regression and classification using multiple feature maps from a CNN network, whereas YOLO performs those using the last feature map from the network [ 34 , 35 ]. Because of these advantages, the SSD model has been widely applied to species classification [ 36 , 37 , 38 ]. However, these studies used the SSD model to classify species belonging to different taxonomic groups, such as genus, family, and order.…”