Image processing method and electronic equipment

The invention provides an image processing method, and the method is characterized in that the method comprises the steps: inputting a to-be-audited object into a content auditing neural network model, and obtaining a plurality of first feature maps outputted by a plurality of convolution layers; performing image up-sampling and image splicing on the plurality of first feature maps to obtain a plurality of second feature maps; performing image fusion on the class activation mapping images corresponding to the plurality of second feature images and the to-be-audited object to generate thermodynamic diagram images or thermodynamic diagram image frames corresponding to the plurality of second feature images; and utilizing the thermodynamic diagram image or the thermodynamic diagram image frame to shield illegal contents in the object to be audited. The invention further provides the electronic equipment. 本申请提供一种图像处理方法，其特征在于，所述方法包括：将待审核对象输入到内容审核神经网络模型，得到多个卷积层输出的多张第一特征图；将所述多张第一特征图分别进行图像上采样和图像拼接，得到多张第二特征图；将所述多张第二特征