An Optimization Method of Deep Transfer Learning for Vegetation Segmentation under Rainy and Dry Season Differences in a Dry Thermal Valley

• Published in: Plants (Basel) Vol. 12; no. 19; p. 3383
• Main Authors: Chen, Yayong, Zhou, Beibei, Ye, Dapeng, Cui, Lei, Feng, Lei, Han, Xiaojie
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 25.09.2023; MDPI
• Subjects: Accuracy; Artificial intelligence; best migration learning conditions; Comparative analysis; Datasets; Deep learning; deep learning optimization; Dry season; dry-hot valley environment; Grasses; Image segmentation; Machine learning; Measurement; Methods; migration training; Network management systems; Networks; Neural networks; Optimization; Plant communities; Rainy season; Reduction; Seasons; Semantic segmentation; Semantics; Task complexity; Training; Transfer learning; Vegetation; vegetation in field; Water conservation; China
• ISSN: 2223-7747; 2223-7747
• DOI: 10.3390/plants12193383

Deep learning networks might require re-training for different datasets, consuming significant manual labeling and training time. Transfer learning uses little new data and training time to enable pre-trained network segmentation in relevant scenarios (e.g., different vegetation images in rainy and dry seasons); however, existing transfer learning methods lack systematicity and controllability. So, an MTPI method (Maximum Transfer Potential Index method) was proposed to find the optimal conditions in data and feature quantity for transfer learning (MTPI conditions) in this study. The four pre-trained deep networks (Seg-Net (Semantic Segmentation Networks), FCN (Fully Convolutional Networks), Mobile net v2, and Res-Net 50 (Residual Network)) using the rainy season dataset showed that Res-Net 50 had the best accuracy with 93.58% and an WIoU (weight Intersection over Union) of 88.14%, most worthy to transfer training in vegetation segmentation. By obtaining each layer’s TPI performance (Transfer Potential Index) of the pre-trained Res-Net 50, the MTPI method results show that the 1000-TDS and 37-TP were estimated as the best training speed with the smallest dataset and a small error risk. The MTPI transfer learning results show 91.56% accuracy and 84.86% WIoU with 90% new dataset reduction and 90% iteration reduction, which is informative for deep networks in segmentation tasks between complex vegetation scenes.