| 基于残差网络和小样本学习的鱼图像识别 |
| |
| 引用本文： | 袁培森,宋进,徐焕良.基于残差网络和小样本学习的鱼图像识别[J].农业机械学报,2022,53(2):282-290. |
| |
| 作者姓名： | 袁培森 宋进 徐焕良 |
| |
| 作者单位： | 南京农业大学 |
| |
| 基金项目： | 国家自然科学基金项目(61502236、61806097)和江苏省农业科技自主创新资金项目(SCX(21)3059) |
| |
| 摘 要： | 针对鱼种类多、数据采集难度大，且需要细粒度图像识别等问题，提出了一种基于度量学习的小样本学习方法。采用基于度量学习的小样本学习网络以及ResNet18的残差块结构，提取鱼图像的深层次特征，并将其映射至嵌入空间进而在嵌入空间判断鱼的种类。为了进一步提升识别准确率，利用小样本学习模型在mini-ImageNet数据集进行预训练，然后将训练的结果迁移到Fish100细粒度数据集上进行精细化训练，得到最终鱼图像识别的判别模型。使用本文模型与常用的5种小样本学习模型，在鱼图像数据集Fish100和ImageNet上进行对比试验，结果表明本文模型的识别效果最佳，在两个数据集上的识别精度分别达到了94.77%和91.03%,且精度、召回率和F1值均明显优于其它模型。
|
| 关 键 词： | 鱼 图像识别 残差网络 小样本学习 迁移学习 |
| 收稿时间： | 2021/2/22 0:00:00 |
Fish Image Recognition Based on Residual Network and Few-shot Learning |
| |
| YUAN Peisen,SONG Jin,XU Huanliang.Fish Image Recognition Based on Residual Network and Few-shot Learning[J].Transactions of the Chinese Society of Agricultural Machinery,2022,53(2):282-290. |
| |
| Authors: | YUAN Peisen SONG Jin XU Huanliang |
| |
| Institution: | Nanjing Agricultural University |
| |
| Abstract: | Accurate and effective identification of fish images play important role in the observation of fish populations and the management of the ecological environment. However, there were some issues, such as lots of kinds of fish, difficulty of data collection in the complex environments, and fine-grained fish image recognition. For solving the problem of few image annotation of fish image, a few-shot learning method based on metric learning was proposed. Firstly, the residual block structure of ResNet18 was used to improve the few-shot learning network based on metric learning, for extracting the deep features of fish images, and then they were mapped to the embedding space for obtaining the mean center by clustering skills. Secondly, for further improving the recognition accuracy, the improved few-shot learning model was used for pre training on the mini-ImageNet dataset, and then the training results were transferred to the Fish100 fine-grained dataset for fine-grained training to get the final discrimination model. Based on this model, comparative experiments were conducted with the existing five few-shot learning models on the fish data set Fish100 and ImageNet. The results showed that the model proposed had the best recognition effect and the recognition accuracy on the two datasets reached 94.77% and 91.03%, respectively, and the accuracy, recall rate, and F1 were significantly better than that of other models. The experiments showed that the method proposed can effectively improve the accuracy of few-shot learning in fish identification with few annotated fish images, which can provide technical support and reference for the application of practical fish image recognition. |
| |
| Keywords: | fish image recognition residual network few-shot learning transfer learning |
| 本文献已被 维普 等数据库收录！ |
| 点击此处可从《农业机械学报》浏览原始摘要信息 |
| 点击此处可从《农业机械学报》下载免费的PDF全文 |
|
