Bacillus thuringiensis (Bt) is widely known for its applications in pest control, the production of genetically modified plants, and controlling the vectors of human diseases. However, new areas of interest have emerged for Bt, such as chitinase production and its potential use against cancer cells. Despite the diversity of applications, some laboratory procedures remain consistent, such as counting Bacillus and Cry proteins ( \(\delta \) -endotoxins). This process has traditionally been conducted manually by humans using microscopy images, which can be challenging, especially when the images are generated by an optical microscope, as in the case of this study. To address this issue, this paper presents digital image processing techniques, called as OIP \(\delta \) C algorithm, to enhance the characteristics of Cry proteins in Bt’s microscopy images, representing the initial stage of automated counting. Despite the difficulty in identifying Cry proteins in images, OIP \(\delta \) C algorithm has yielded promising results that can be utilized by machine learning algorithms for protein counting.