融合多尺度特征与多列卷积网络的水稻苗无人机遥感计数方法

Rice Seedling Counting Method Based on Multi-column Convolutional Network Integrating Multi-scale Features

摘要:

水稻早苗期种植密度与产量估算、田间管理密切相关。随着计算机视觉技术的发展, 利用无人机图像识别技术自动统计水稻数量正逐步取代传统的人工采样和计数方法。然而, 无人机拍摄的图像存在水稻苗形状不规则、水稻苗重叠、尺度变化和背景遮挡等问题, 导致特征学习出现偏差, 进而影响了水稻密度的准确估算和计算预测的效率。水稻苗之间显著的相对大小差异也成为影响计数精度的另一重要因素。针对这些问题, 提出了一种基于改进的多列卷积神经网络的水稻苗计数算法(Rice counting network, RCN)。该算法根据目标的大小和形状, 提出了改进高斯密度核, 结合水稻图像与中心点RGB图像的相似度信息，使密度核更加贴合水稻苗纹理。对不同大小和形状的稻穗数量进行量化, 并对尺度进行精细划分, 计算出需要融合的不同特征感受野大小，采用融合膨胀卷积模块，增加了感受野大小，增强了模型对多尺度环境中的处理效果。在此基础上, 对不同尺度的通路添加合适的注意力机制, 增强不同尺度下对水稻苗合理信息的关注, 在复杂背景、多尺度和密集遮挡的场景中体现出了明显的有效性。在高度5m捕获的RSD2023数据集上, RCN的平均绝对误差为3.10穴/m2, 均方误差为4.03穴2/m4，优于MCNN、CSRNet、SCAR和TasselNetV2等主流算法。

Abstract:

The planting density of rice in the early seedling stage is closely related to yield estimation and field management. With the development of computer vision technology, the use of drone image recognition technology to automatically count rice is gradually replacing traditional manual sampling and counting methods. However, images taken by drones have problems such as irregular rice seedling shapes, overlapping rice seedlings, scale changes, and background occlusion, which lead to deviations in feature learning, which in turn affects the accurate estimation of rice density and the efficiency of computational prediction. Significant relative size differences among rice seedlings will also become another important factor affecting counting accuracy. In response to these problems, a rice seedling counting algorithm (rice counting network, RCN) was proposed based on an improved multi-column convolutional neural network. Based on the size and shape of the target, an improved Gaussian density kernel was proposed, which combined the similarity information of the rice image and the RGB image of the center point to make the density kernel more suitable for the texture of the rice seedlings. The number of rice ears of different sizes and shapes was quantified, and the scales were divided finely to calculate the receptive field sizes of different features that needed to be fused. The fusion expansion convolution module was used to increase the receptive field size and enhance the model’s multi-scale capabilities. On this basis, appropriate attention mechanisms were added to the pathways at different scales to enhance attention to reasonable information about rice seedlings at different scales. It showed obvious effectiveness in scenes with complex backgrounds, multi-scales and dense occlusions. On the RSD2023 data set captured at a height of 5 m, RCN’s mean absolute error MAE=3.10holes/m2 and mean square error MSE=4.03holes2/m4, which were better than that of the mainstream algorithms such as MCNN, CSRNet, SCAR and TasselNetV2