Research On Eye-glasses Refractive Parameters Measurement System

Vision is the major means for human beings to perceive the world and obtaininformation. Modern society lifestyles including the popularity of TV, the application ofvideo and the networking of information bind people tightly in front of screens. Accordingly,visual problems occur easily and frequently due to using eyes for a long time as well as poorconsciousness of self prevention and protection. People demand more and more in their visualquality as well as eye health and protection. However, the existing service resources andquality of visual correction is very limited in China. Taking vision correction for example, asmuch as41%of Chinese juveniles aged5to15have refractive errors such as myopia,hyperopia and astigmatism, but85%of them have not been effectively corrected. Theproportion is much higher than that of the countries in Europe and other regions. One of themain reasons for this situation is that the existing studies on domestic refraction measurementdevices are relatively weak, which leads to my research subject, namely “Research onEye-Glasses Refractive Parameters Measurement System”.The research is based on the theories in visual optics, glasses optics and computerscience. With optometric measurement devices as the object of study, this paper analyzes thedifference between the existing domestic devices and foreign devices in measurementtechnology and conducts the following studies on the changing rule of optical paths of eye orglasses with different refractive power in eye or glasses measurement systems as well as theestablishment of corresponding mathematical models:(1) A study on human cornea parameters measurement and human eye locationtechnology is conducted. It analyzes the existing algorithm which is based on the geometricfeatures location and the measurement of human cornea parameters, points out theinadaptability of this algorithm towards the measurement of corneas with astigmatism, anddeduces a mathematical expression of adding cornea astigmatism correction ΔD on the basisof the existing algorithm. Meanwhile, it puts forward a new algorithm based on the variancerate of image grayscale to locate human eyes. By analyzing the grayscale variance characteristics of the defocus and the corneal reflection images obtained from the focal plane,this algorithm adopts dual thresholds grayscale image integration method to calculate thevariance rate of image grayscale K, and deduces formulas to calculate K, the distance ofdefocus D, and the correction value of corneal curvature radius ΔR. The experimental resultsshow that both algorithms reach the requirements of national standards with the measurementerror of less than±0.02mm when they are used to measure corneal curvature model eye.(2) Compared with the algorithm based on the geometric features location and themeasurement of human cornea parameters, the algorithm based on grayscale variance rate forlocation has more advantages in the difficulty of image processing and computation speed.Therefore, a study on the three-dimensional tracking system of the algorithm based ongrayscale variance rate is conducted in the experimental prototype.(3) This paper studies the calculation method of human eye refractive parameters basedon retinal reflection images, and establishes mathematical models of human eye diopter andretinal reflection images as well as optical compensation system displacement. Besides, itfurther studies the fundamental of the existing grayscale image skeleton extraction algorithmand its inadaptability, and proposes an adaptive skeleton extraction algorithm for dynamicgrayscale thresholds. The algorithm can be used for binarization processing by adoptingdifferent thresholds in different areas with the same image; the setting of thresholds is onlyrelated to the grayscale distribution of relevant neighborhood areas; it judges the skeletonpoints and extracts the skeletons according to the grayscale distribution characteristics, thusavoiding the loss of image features. The experimental results show that the algorithm has theadvantages of complete extraction of image features and fast computation speed, and that itcan effectively improve the accuracy of anisometropia measurement related to human eyes.(4) The paper studies the combinational algorithm of spherical lens and cylinder lens inthe subjective auto refractor measurement system by using NIDEK3100(an autorefractor)as the research platform, and sets up a hardware control system to conduct furtherresearch on the control principle of auto refractor.(5) The paper discusses the fundamental for measurement of the lens vertex power, focuses on the optical paths of different glasses designs on the lensmeter, and describes thecalculation method of the lens vertex power.