Research On Low Quality Fingerprint Recognition Combined With Iris Recognition

In the border security screening system,the identification of drivers has always been a difficult problem to be solved.At present,most of the system identification methods choose fingerprint identification technology.Because of its earliest technology development and the highest maturity,most drivers' fingerprints belong to low-quality fingerprint images due to the nature of work.The fingerprints of fracture,blur,and defect are intricate.The traditional single-modal biometric identification technology is largely limited to a single application field.In the face of this situation,this paper proposes an algorithm for low-quality fingerprint recognition combined with iris recognition..Considering the driver's visually good and high iris recognition accuracy,this paper will combine the low-quality fingerprint and iris recognition technology to identify the driver.As technology continues to evolve,multiple biometric methods combined with verification techniques are technically supported and will be relevant to our daily lives in the near future.The main work of this paper is as follows.1.In-depth understanding and research on the theoretical knowledge of biometric technology in four data layers,and compare the advantages and disadvantages of each data layer,combined with the strong application of border control system,this paper proposes a fusion scheme of fingerprint and iris recognition technology in the matching layer.Matching layer fusion is to optimize the system and improve the accuracy and enhance the system performance through the matching matching layer fusion algorithm after each single-mode biometric identification completes the matching work independently.2.In the face of the driver's low-quality fingerprint,eight different types of low-quality fingerprints were selected from the FVC2000 database to simulate the driver's fingerprint.Two fingerprint identification methods were proposed.The data were analyzed to compare the recognition effects of the two methods.In the first method,the DCT dimension reduction processing step is introduced,and the data energy is concentrated to the upper left corner by the zigzag transformation,which greatly improves the recognition speed.The second method is based on fingerprint repair.The fingerprint information points are facilitated by a series of refinement steps.3.In the iris recognition system,using the low-quality iris image as the input data,a more accurate two-step method of iris localization is proposed,that is,the canny operator combined with the Hough transform localization method,and the high-accuracy algorithm for the eyelid and the eyelash part is performed,and then The data is normalized and enhanced,and the matching score is finally output.4.Fingerprint recognition technology and iris recognition technology in the matching layer fusion,this paper firstly through the simple mean fusion method,the experimental data can be deeply understood that although the recognition system performance is improved than the single-modal biometric technology,but its right The value is fixed and cannot meet certain application requirements.Therefore,this paper adjusts the weight distribution,and uses the genetic optimization method to give the best weight distribution and significantly improve the system recognition rate.