Image processing presents an intriguing realm with both challenges and opportunities, inviting researchers to devise innovative models that streamline processing tasks. Data acquisition stands as a pivotal element in numerous image processing models, incorporating techniques like deblurring and noise removal. Additionally, these models contend with occlusion factors stemming from alterations in the input image’s lighting conditions. Several authors have delved into investigating this issue to identify a solution. Their efforts have involved manual intervention, although a limited number of authors have sought to minimize human involvement. The detection of shadows in scanned documents remains a relatively unexplored area. This paper introduces a novel model designed to identify shadows in scanned documents, leveraging the YCbCr color space components, primarily focusing on the illumination aspect. The proposed model is capable of detecting shadows automatically, avoiding manual intervention, which distinguishes from previous models. This approach streamlines the shadow detection process, and the proposed model attains an impressive accuracy when tested on a dataset comprising 350 images crafted to mimic real-world conditions. However, it’s worth noting that the model’s performance may be hindered in non-shadow regions due to consideration of illumination factors in the shadowed area.