Image Sampling Based on Dominant Color Component for Computer Vision

Image sampling is a fundamental technique for image compression, which greatly improves the efficiency of storage, transmission, and applications. However, existing algorithms primarily consider human visual perception discard irrelevant information based on subjective preferences. Unfortunately, these methods may not adequately meet demands computer vision tasks can even lead to redundancy because different preferences between computer. To tackle this issue, paper investigates key features vision. Based our findings, we propose an method dominant color component (ISDCC). In method, utilize grayscale preserve essential structural Then, construct concise feature map channel pixels. This approach provides relevant tasks. We conducted experimental evaluations using well-known benchmark datasets. The results demonstrate that ISDCC adapts effectively requirements, significantly reducing amount data needed. Furthermore, has minimal impact performance mainstream across various Compared other approaches, proposed exhibits clear advantages by achieving superior with less usage.