Digital Capsule Endoscopy System And Image Interpolation Algorithmic Research

Gastrointestinal disease is increasingly becoming a major threat to health.Traditional gastrointestinal disease detection means such as gastroscopy,colonoscopy,intrusive inside the body,brings great suffering to the patients,and range of detection is limited.There is an urgent need to improve.With the development of microelectronic technology,information technology,capsule endoscope become a new gastrointestinal disease means for diagnosis and treatment.Capsule endoscope brings smaller pains,can detect the whole digestive tract,and images are clear and stable.Therefore,capsule endoscope is an important research subject in the direction of gastrointestinal diseases.Given imaging company in Israel,put forward the first examination of gastrointestinal endoscope system in 2001,namely the M2 A system.This is the first time.Its size is 11 mm diameter x 30 mm length,weighing about 3.7 g,relying on the motility of the gastrointestinal tract to move forward in the gastrointestinal tract,and can work for 6 to 8 hours,shoot images of the inner wall of the small intestine.This method eliminates the traditional methods of bringing pain and side effects,eliminates the dead angle of traditional detection methods,also can detect the whole digestive tract.This has been widely welcomed.Since then,many countries around the world have also carried out research on this type of testing equipment.Traditional capsule endoscope endoscopy collects and transmits analog image.With weak anti-interference ability,low frame rate,and low resolution,there are still much room for improvement.This paper is funded by the National Natural Science Fund Project(project number: 31170968),Shanghai Municipal Science and technology project(project number: 14441902800 15441903100.In this paper,according to the limitation and insufficiency of traditional analog capsule endoscope endoscopy,we put forward a kind of improved digital capsule endoscope system.This system adopts a wireless power supply and image capsule using new highresolution image sensor OV6930 and analog to digital converter OV420,which can collects a frame rate of 30 frames per second,and resolution for 400 × 400 digital image.The capsule transmits digital signal,with good antiinterference ability,and digital image good for subsequent image processing,such as pattern recognition,intelligent detection.There are several specific improvements:1.The use of white light and ultraviolet light as the light source.White light and ultraviolet light turn on in turn,while white light is used for routine detection,and after the discovery of abnormal sites using ultraviolet light to detect whether there are early carcinogenesis,and determine the form of cancer.2.Adopting a new type of image sensor OV6930 and analog to digital converter OV420.Get a ten bit digital image,and the resolution is 400 × 400.The frame rate is 30 frames per second.Between the sensor and the analog to digital converter,and between the MCU and the analog to digital converter,the SCCB protocol is used to communicate.3.In this paper,the design of the capsule endoscopy system,takes a series of measures to reduce data volume: image compression in the internal capsule,image interpolation done outside body.4.Aiming at the characteristics of the gastrointestinal tract image,this paper optimizes the JPEG-LS algorithm and achieves the expected compression ratio.The main optimization methods,such as reduce the pixel length encoded entry criteria,ignoring each line at the beginning and end of the tour,the predictor formula optimization,expand the range of pixels using the predictor and.After optimization,the algorithm achieves the expected compression ratio.5.A new image interpolation method is presented in this paper,through the use of the original image correlation between different color planes,the interpolated image obtained by the method of weighted mean is loop corrected using less amount of calculation,to obtain color image of high quality.The experimental data show that color images of the algorithm has a high quality,so we have achieved the desired purpose.