融合迁移学习和集成学习的自然背景下荒漠植物识别方法
［目的/意义］荒漠植物的准确识别是其认识和保护过程中不可或缺的任务，是荒漠生态研究与保护的基础。自然条件下野外荒漠植物图像的机器视觉自动分类识别可有效提升植物资源调查效率、降低人为主观因素影响，对荒漠植物的精准分类、多样性保护和资源化利用具有重要意义。［方法］以自然环境下的整株荒漠植物图像为研究对象，构建新疆干旱区荒漠植物图像数据集，以EfficientNet B0—B4网络为基础网络，提出一种融合迁移学习和集成学习的荒漠植物图像识别算法，并在公开数据集Oxford Flowers102上进行对比验证。［结果和讨论］基于EfficientNet B0网络的单一子模型的 Top-1准确率最高可达 93.35%，最低为 92.26%，软投票Ensemble-Soft 模型、硬投票 Ensemble-Hard 模型以及加权投票法集成的 Ensemble-Weight 模型的准确率分别为 93.63%、93.55%和 93.67%，F1 Score 和准确率相当；基于 EfficientNet B0—B4 网络的单一子模型的 Top-1 准确率最高可达 96.65%，F1 Score 为96.71%，而 Ensemble-Soft 模型、Ensemble-Hard 模型以及 Ensemble-Weight 模型的准确率分别为 99.07%、98.91%和99.23%，相较于单一子模型，精度进一步提高，F1 Score 与准确率基本相同，模型性能显著；在公开数据集OxfordFlowers102上进行对比试验，3个集成模型相比 5个子模型准确率和F1 Score 最高提升了 4.56%和 5.05%，最低也提升了 1.94%和 2.29%，证明了本研究提出的迁移和集成学习策略能够有效提高模型性能。［结论］ 本方法可提高荒漠植物的识别准确率，通过云端传输至服务器后，实现荒漠植物的准确识别，为真实野外环境下植物图像识别精度低、模型鲁棒性及泛化性弱等问题提供解决思路。服务于野外调查、教学科普以及科学实验等场景。
环境生物学植物学计算技术、计算机技术
荒漠植物识别自然背景集成学习迁移学习投票法数据集
孙伟,王亚鹏,李全胜,曹姗姗.融合迁移学习和集成学习的自然背景下荒漠植物识别方法[EB/OL].(2023-08-14)[2025-10-29].https://chinaxiv.org/abs/202308.00171.点此复制
[Objective] Desert vegetation is an indispensable part of desert ecosystems, and its conservation and restoration are crucial. Accurateidentification of desert plants is an indispensable task, and is the basis of desert ecological research and conservation. The complexgrowth environment caused by light, soil, shadow and other vegetation increases the recognition difficulty, and the generalization ability is poor and the recognition accuracy is not guaranteed. The rapid development of modern technology provides new opportunitiesfor plant identification and classification. By using intelligent identification algorithms, field investigators can be effectively assistedin desert plant identification and classification, thus improve efficiency and accuracy, while reduce the associated human and materialcosts.[Methods]In this research, the following works were carried out for the recognition of desert plant: Firstly, a training dataset of deeplearning model of desert plant images in the arid and semi-arid region of Xinjiang was constructed to provide data resources and basicsupport for the classification and recognition of desert plant images.The desert plant image data was collected in Changji and Tachengregion from the end of September 2021 and July to August 2022, and named DPlants50. The dataset contains 50 plant species in 13families and 43 genera with a total of 12,507 images, and the number of images for each plant ranges from 183 to 339. Secondly, a migrationintegration learning-based algorithm for desert plant image recognition was proposed, which could effectively improve the recognitionaccuracy. Taking the EfficientNet B0B4 network as the base network, the ImageNet dataset was pre-trained by migrationlearning, and then an integrated learning strategy was adopted combining Bagging and Stacking, which was divided into two layers.The first layer introduced K-fold cross-validation to divide the dataset and trained K sub-models by borrowing the Stacking method.Considering that the output features of each model were the same in this study, the second layer used Bagging to integrate the outputfeatures of the first layer model by voting method, and the difference was that the same sub-models and K sub-models were comparedto select the better model, so as to build the integrated model, reduce the model bias and variance, and improve the recognition performanceof the model. For 50 types of desert plants, 20% of the data was divided as the test set, and the remaining 5 fold cross validationwas used to divide the dataset, then can use DPi(i=1,2,,5) represents each training or validation set. Based on the pre trained EfficientNetB0B4 network, training and validation were conducted on 5 data subsets. Finally, the model was integrated using soft voting, hard voting, and weighted voting methods, and tested on the test set.[Results and Discussions]The results showed that the Top-1 accuracy of the single sub-model based on EfficientNet B0 networkwas 92.26%~93.35%, the accuracy of the Ensemble-Soft model with soft voting, the Ensemble-Hard model with hard voting and theEnsemble-Weight model integrated by weighted voting method were 93.63%, 93.55% and 93.67%, F1 Score and accuracy were comparable,the accuracy and F1 Score of Ensemble-Weight model integrated by weighted voting method were not significantly improvedcompared with Ensemble-Soft model and Ensemble-hard model, but it showed that the effect of weighted voting method proposed inthis study was better than both of them. The three integrated models demonstrate no noteworthy enhancements in accuracy and F1Score when juxtaposed with the five sub-models. This observation results suggests that the homogeneity among the models constrainsthe effectiveness of the voting method strategy. Moreover, the recognition effects heavily hinges on the performance of the EfficientNet B0-DP5 model. Therefore, the inclusion of networks with more pronounced differences was considered as sub-models. A single sub-model based on EfficientNet B0B4 network had the highest Top-1 accuracy of 96.65% and F1 Score of 96.71%, while Ensemble-Soft model, Ensemble-Hard model and Ensemble-Weight model got the accuracy of 99.07%, 98.91% and 99.23%, which furtherimproved the accuracy compared to the single sub-model, and the F1 Score was basically the same as the accuracy rate, and the modelperformance was significant. The model integrated by the weighted voting method also improved accuracy and F1 Score for both softand hard voting, with significant model performance and better recognition, again indicating that the weighted voting method wasmore effective than the other two. Validated on the publicly available dataset Oxford Flowers102, the three integrated models improvedthe accuracy and F1 Score of the three sub-models compared to the five sub-models by a maximum of 4.56% and 5.05%, and aminimum of 1.94% and 2.29%, which proved that the migration and integration learning strategy proposed in this paper could effectivelyimprove the model performances.[Conclusions]In this study, a method to recognize desert plant images in natural context by integrating migration learning and integrationlearning was proposed, which could improve the recognition accuracy of desert plants up to 99.23% and provide a solution tothe problems of low accuracy, model robustness and weak generalization of plant images in real field environment. After transferringto the server through the cloud, it can realize the accurate recognition of desert plants and serve the scenes of field investigation, teachingscience and scientific experiment.
desert plant image classificationnatural backgroundensemble learningtransfer learningvoting methoddataset
展开英文信息
评论