Autofocus microscope system based on blur measurement approach

Abstract

Abstract Blurred (out-of-focus) images hinder the process of accurate cell counting in medical images. Cell counting is considered an important method to quantify cells for disease diagnosis and medical research. The hemocytometer is used to count cells as it is a cheaper and simple method. The specimen which contains cells is loaded into the hemocytometer and placed on a microscope. In this paper, a new autofocus technique is proposed to automatically adjust the microscope knobs by controlling the movement of adjusting knobs. This method is based on measuring and evaluation of the amount of blur annoyance in cell images. Whereas, adjusting knobs of microscope is a time consuming especially for beginners. Also, the automatic adjustment avoids losing the knob gear if it is used for several times or used in a wrong way. A CCD camera is attached to the microscope to capture images of specimen under test. An efficient blur metric is used to evaluate the amount of blur in captured images. The microscope knob is attached to a stepper motor to control its movement. The movement is stopped when the well-focused image is obtained. The proposed method has proved that it is an efficient method to automatically adjust the microscope knobs and get a well-focused image. The proposed method is compared with another recent approach that is based on an evaluation of a contrast measurement in images to automatically control the microscope knobs. The results proved that the output in-focus images by using the proposed approach for auto-focusing the microscope system have low blur values and much closer to the quality of the original reference images. Moreover, the proposed method which is based on blur measurement gives a higher image quality than one of the most recent approaches that depends on contrast measurement.