Abstract

Experiments on animals is a key step in scientific and medical studies. In some experiments on animals, mouse motions is studied in a place called open field maze. Until now, track and analysis of mouse motions has been done by hand or computer. Todays, manual method for mouse motions analysis is not utilized due to its time cost and low accuracy. Meanwhile, in a computer aided analysis method, video images are captured with a digital camera and then mouse motions analysis is facilitated with image processing algorithms and a graphical user interface. In this project, design and implementation of an application software will be done for mouse motions analysis in an open field maze. In this project, activities are classified into two group: hardware section and software section. Hardware section includes the construction of open field maze and also the selection of needed capturing equipment for video images and transmission of data to a computer. Software section has subsections called image processing algorithms and graphical user interface design. After capture and transmission video images into computer, it is necessary to be processed these images for extract mouse motions information. This step on its own have two subsections: calibration and mouse motion information extraction. Calibration stage is done before any further test with a mouse. There is no need to calibration in remaining trials on that work session. Mouse motion information extraction stage in every frame of input video images is pursued as the following: 1) color image conversion to a gray level image, 2) image filter for noise reduction, 3) image thresholding to reach a binary image, and 4) mouse centroid estimation in binary image.  To simple and efficient interaction of a researcher with the software, we forced to design and implement a graphical user interface with the help of MATLAB App designer. This GUI allows access to video images, calibration section, and mouse motions analysis section. Also, MATLAB App provide the visualization possibility of mouse tracking as a diagram. In this paper, tracking results was provided on a mouse. Animal laboratory of ZMUS (Zanjan University of Medical Sciences) allowed us to test our tracking system on a mouse. It is worth to mention that developed software in our research work is not sensitive to mouse size. After many studies, we designed the preliminary version of graphical user interface to have the following sections: Video source, Arena, Researcher, Trails, Calibration, View, and report. In the present research work, we were able to design and implement GUI to analysis mouse motions in open field maze. The software is capable to execute experiments and store mouse motions data.