Day and night universal image processing method, device and equipment based on silicon sensor camera

The invention provides a day and night universal image processing method, device and equipment based on a silicon sensor camera, and the method comprises the steps: obtaining a to-be-processed original image which is a mixed image of visible light and near infrared light; processing the original image by adopting a pre-trained neural network model in a separation stage to obtain a visible light image; and adopting a pre-trained neural network model in a restoration stage to restore the visible light image to obtain a restored color image, thereby achieving image enhancement and obtaining a high-quality visible light image. 本申请提供一种基于硅传感器相机的昼夜通用图像处理方法、装置及设备，该方法包括：获取待处理的原始图像，其中，所述原始图像为可见光和近红外的混合图像；采用预先训练好的分离阶段的神经网络模型对所述原始图像进行处理，获得可见光图像；采用预先训练好的修复阶段的神经网络模型，对所述可见光图像进行修复处理，获得修复后的彩色图像，实现了图像增强，得到高质量的可见光图像。