This study aims to extract and analyze the color of urban vitality in street view images based on deep learning methods. With street view image data collected from multiple different cities, we performed detailed preprocessing steps, including cropping, scaling, normalization, and data augmentation, to standardize the image input format. Using a convolutional neural network (CNN) model, we successfully extracted color features from the image and revealed the color structure of urban vitality through color histograms and HSV color feature distributions. Experimental results show that there is a significant correlation between color features and urban vitality, and high color saturation and brightness usually correspond to higher urban vitality. The research conclusions demonstrate the effectiveness and practicality of the deep learning-based street view image color feature extraction method in urban planning and design.