| Summary: | Leukemia is a type of cancer that affects the white blood cell. Early detection of leukemia is important to reduce the rate of mortality. In order to detect acute leukemia, conventional screening method based on microscopic image is used, where sample of blood cell will be taken from the suspected leukemia patient and manually white blood cell (WBC) condition is observed using microscope. The manual screening process is tedious, time consuming and usually prone to error due to low contrast between the nucleus and cytoplasm of WBCs. This report introduces a new enhancement method which is a combination of Particle swarm optimization (PSO) algorithm and contrast stretching, known as Hybrid PSO-Contrast stretching (HPSO-CS). The PSO has been used to optimize the fitness criterion in order to improve the contrast and detail in microscopic image by adapting the parameters as a contribution to enhancement technique. In this study, PSO algorithm is used to perform image segmentation to remove all the unwanted part such as red blood cell (RBC), platelet and also the background while retain the WBC part. The segmentation algorithm uses saturation S-component based on Hue, Saturation, Intensity (HSI) color model. After the segmentation is done, contrast stretching process is applied to the original image to stretch intensity of the pixel. Then the segmented image is combined with the resultant image that has been stretched to produce the enhanced image. The results of the proposed method are evaluated by using mean-square error (MSE), Peak-signal-to-noise-ratio (PSNR) and Absolute mean brightness error (AMBE). This proposed method is benchmarked by comparing against two image enhancement methods, global enhancement and Class Limited Adaptive Histogram Equalization (CLAHE). Based on the results, it can be concluded that quality of the enhanced image for the proposed method is much better with the lowest MSE (2067.651), AMBE (43.51827) and highest PSNR (14.98671) compared to th...
|
|---|