Sputum smear image enhancement for tuberculosis detection with HSV color space and histogram equalization

Abstract

Tuberculosis (TB) is an infectious disease that mainly affects the lungs and is still a major public health concern around the world. For detecting tuberculosis infection, the Ziehl-Neelsen staining technique is commonly used. The manual screening process is time-consuming and labor-intensive. To boost the screening process, image enhancement methods may be used. The problems get much worse if the captured images are blurred or have low contrast. Image enhancement can be used to reveal information that was previously obscured as well as to improve the image contrast and the accuracy of images. The combination of filtering, color space conversion, and histogram equalization techniques are proposed in this paper to improve the quality and the accuracy of images for TB bacilli detection. This approach enhances the quality of the sputum image while maintaining accuracy. Furthermore, this approach results in a greater color contrast between the bacteria and the picture background which increases the performance of the segmentation process.