Research On The Measurement Of MTF In Video Endoscope By Knife-edge Method

Medical endoscope is an essential diagnostic instrument in the medical field,and the doctor can directly observe tissue organs in the human body through it.The quality of endoscopic imaging can directly affect the patient's life and safety,causing serious medical accidents.Endoscopic performance indicators of the international standard is lagging behind,has been unable to meet the needs of the development of the times.Therefore,this paper uses the knife-edge method to effectively measure the MTF of the endoscope,hoping to introduce a more scientific optical transfer function to the endoscopic evaluation system.Medical endoscopes are characterized by large distortion,wide field of view and large depth of field,and can not be measured in the existing modulation transfer function(MTF)measurement device.In domestic for endoscope image quality evaluation is still on the resolution of the evaluation,medical endoscope measurement of modulation transfer function was still in the stage of development.In this paper,we designed the knife-edge method to measure the video endoscope,and then use the knife-edge scanning method and the knife-edge imaging to obtain the knife-edge diffusion function(ESF),then taking the derivative and Fourier transform of Edge Spread Function to calculate the Modulation Transfer Function of endoscope.In this paper,the automatic measurement of medical endoscopic video modulation function is realized by the knife-edge scanning method and the knife-edge imaging.These two different knife-edge method have their unique characteristics,which can meet different requirements.However,the precision of the knife-edge imaging of this system is low and not recommended.The nonuniformity of the light source is compensated for the accuracy of the measurement when the knife-edge imaging is used.Face shot noise produced by the CMOS and the environment of random noise,this paper,comparing the different smoothing filtering method and choose the most conforms to the way of filtering.At the same time,this experiment also improves and optimizes the filtering algorithm,fast Fourier transform algorithm and image processing algorithm,and adds some new functions such as uniformity detection of light source,one-button compensation of light source and automatic capture center image.