Less is more when applying transfer learning to multi-spectral data

Transfer Learning is widely recognized as providing incredible benefits to many image processing tasks. But not all problems in the computer vision field are driven by traditional Red, Green, and Blue (RGB) imagery as tend to be assumed by most large pre-trained models for Transfer Learning. Satellite based remote sensing applications for example typically use multispectral bands of light. While transferring RGB features to this non-RGB domain has been shown to generally give higher accuracy than training from scratch, the question remains whether a more suitable fine tuned method can be found. Given this challenge, this paper presents a study in multispectral image analysis using multiple methods to achieve feature transfer. Specifically, we train and compare two pre-trained models based on the Resnet50 architecture and apply them to a multi-spectral image processing task. The key difference between the two models is that one was pre-trained in a conventional way against RGB data, while the other was trained against a single band greyscale variant of the same data. Our results demonstrate an improved performance on the greyscale pre-trained model relative to the more traditional RGB model.