Realization And Research Of Real-time Image Detection In Corneal Topograph

Eye is an important organ of human to access external optical information.The structure of the eye which is complex,once the eye has been lesions,it is necessary to rely on optical instruments to check it.Corneal topography as an optical measuring instrument for detecting human eye lesions,it can be used to quantitatively detect the surface morphology of human cornea.And at the same time,the topography information of the corneal surface can be obtained by data or processed images.The result have great significance for the diagnosis,treatment and prognosis of corneal optical lesions.Currently,there are many kinds of corneal topography products based on different measurement methods.The Common methods of measurement include Placido disk method,stereo triangular grating grid projection method,optical tomography coherence imaging method,corneal curvature method,etc.The corneal topography based on Placido disc is widely used to accurately display the corneal surface morphology,which provides a powerful means for theophthalmologist to measure the corneal surface morphology and corneal refractive correction or surgical treatment.This paper introduces the real-time image detection of corneal topography.According to the requirement of image acquisition and transmission of corneal topography,the driving system of CCD and the data acquisition and transmission system of USB2.0 are designed.Combined with the requirement of spectral response,transmission frame frequency,resolution and data transmission of corneal topography imaging system,IXC445 ALA is selected as the sensor of image acquisition system.According to the driving timing requirement,the design of drive circuit is realized by AD conversion,CPLD timing frame and single chip microcomputer timing control.By configuring the internal register of AD,the acquisition of image digital signal is realized,and the data communication between PC and USB2.0 is realized by using the firmware driver of USB2.0 and the control of single chip microcomputer.At the same time,in order to realize real-time image detection in corneal topography,this paper based on the optical imaging system in corneal topography,the influence of system defocusing on image sharpness is analyzed.and an image definition evaluation method is proposed.In this method,Laplace is chosen as the evaluation parameter of the coarse tone of the global image,and the quadratic Gaussian function is used to preprocess all kinds of noises,including thenoise data from real eyelashes.Then,the keratogram obtained by Hoff circle transform is introduced.Finally,the edge detection method of Mixed CanBel is used to find out the clear pixel edge,and the clear Placido disk image is obtained in real time.Because the images projected by the Placido disk in the human eye are a group of concentric rings with equal spacing composed of 24 rings,the center point is determined by the ring contour,and the center coordinates of the subpixel level are obtained by adjusting the center point position by the subpixel edge.In order to reconstruct the corneal surface and process data,it is very important to process the Placido image and get the accurate data of each ring.Due to the operation characteristics of such corneal topographic system,the consecutive images’ centers are unavoidable drifted during the process of image acquisition.So it is necessary to match the same serial target pictures.According to the above characteristics,this paper uses the two-dimensional Gaussian smoothing function to preprocess the noise data of all kinds of noises,including real human eyelashes removing and uses mixed Canbel and regional three-polynomial fitting method to effectively detect the sub-pixel edges of corneal images which we derive.The 256 interpolation data per ring is obtained by fitting method.Based on the experiments of standard simulation eye and real human eye,it is proved that the algorithm presented in this paper has good robustness and detection accuracy(the refractive error distributions ofstandard simulation eyes with different curvatures are less than 0.25D),which meets the operation requirements of actual systems used in industry.So our algorithm can be used in such corneal topographer in order to help detection of various eye diseases.