Wang Xianfeng, Zhang Chuanlei, Zhang Shanwen
School of Information Engineering, Xijing University, Xi'an 710123, China.
College of Artificial Intelligence, Tianjin University of Science and Technology, Tianjin, 300222, China.
Comput Intell Neurosci. 2021 Jul 5;2021:5529905. doi: 10.1155/2021/5529905. eCollection 2021.
Plant species recognition is a critical step in protecting plant diversity. Leaf-based plant species recognition research is important and challenging due to the large within-class difference and between-class similarity of leaves and the rich inconsistent leaves with different sizes, colors, shapes, textures, and venations. Most existing plant leaf recognition methods typically normalize all leaf images to the same size and then recognize them at one scale, which results in unsatisfactory performances. A novel multiscale convolutional neural network with attention (AMSCNN) model is constructed for plant species recognition. In AMSCNN, multiscale convolution is used to learn the low-frequency and high-frequency features of the input images, and an attention mechanism is utilized to capture rich contextual relationships for better feature extraction and improving network training. Extensive experiments on the plant leaf dataset demonstrate the remarkable performance of AMSCNN compared with the hand-crafted feature-based methods and deep-neural network-based methods. The maximum accuracy attained along with AMSCNN is 95.28%.
植物物种识别是保护植物多样性的关键步骤。基于叶片的植物物种识别研究既重要又具有挑战性，这是因为叶片在类别内部差异大、类别之间相似度高，而且存在丰富的大小、颜色、形状、纹理和脉络不一致的叶片。大多数现有的植物叶片识别方法通常将所有叶片图像归一化到相同大小，然后在单一尺度上进行识别，这导致性能不尽人意。本文构建了一种新颖的带注意力机制的多尺度卷积神经网络（AMSCNN）模型用于植物物种识别。在AMSCNN中，多尺度卷积用于学习输入图像的低频和高频特征，并且利用注意力机制来捕捉丰富的上下文关系，以实现更好的特征提取并改进网络训练。在植物叶片数据集上进行的大量实验表明，与基于手工特征的方法和基于深度神经网络的方法相比，AMSCNN具有卓越的性能。AMSCNN所达到的最高准确率为95.28%。
