基于卷积神经网络的沿海水质评价技术研究

Research of Coastal Water Quality Evaluation Technology Based on Convolutional Neural Network

针对沿海水质评价,利用遥感图像获取的实时与大范围优势,提出基于卷积神经网络的沿海水质综合评价方法。该方法应用非线性回归模型对遥感图像进行校正;建立基于卷积神经网络的沿海水质评价模型;引入水质评价先验知识,结合遥感图像形成多模态矩阵输入数据;通过多层卷积与池化操作,降低因图像平移、缩放、倾斜等变换引起的误差,提高水质评价精度。实验结果表明,该方法可较准确地评价沿海水质,具有一定的实用价值。

Aim Aiming at the evaluation of coastal water quality,using the real-time and large-scale advantage of remote sensing image acquisition, a comprehensive evaluation method of coastal water quality based on convolutional neural network is proposed. The satellite remote sensing image are corrected by applying nonlinear regression model, and the coastal water quality evaluation model based on convolutional neural network is established. The multimode matrix input data is formed by Adding the prior knowledge of water quality evaluation and combining the remote sensing image, and through the multi-layer convolution and pooling operation, it reduce the error caused by the image translation, scaling, tilting and so on, and improve the water quality evaluation accuracy. The experimental results show that the method can accurately evaluate the coastal water quality and has a practical value.