透過您的圖書館登入
IP:18.118.146.60
  • 學位論文

使用CPLD強化鈔票影像之對比與雜訊之消除

The Contrast Enhancement and Noise Reduction of Banknote Image by Using CPLD

指導教授 : 駱榮欽

摘要


本研究旨在發展一台加強型多槽式鈔票整理機,其功能包含鈔票驗鈔、點鈔、影像特徵辨識以及整合這些功能的人機介面。整體系統包含馬達速度控制電路、鈔票厚度辨識電路、影像強化輸入介面、主控電路版Advanced RISC Machine-11(ARM-11)硬體規劃、周邊相關電路。論文主要研究由接觸式影像感測器讀取鈔票類比信號,經改善的硬體電路放大類比信號並轉換成八位元灰階影像數位信號,再經由Complex Programmable Logic Device(CPLD) MAXII做硬體即時影像強化處理,使讀取的影像可以改善對比並強化影像特徵點減少雜訊干擾。首先影像信號經放大器倍率調整改善及使用CPLD強化影像對比,影像強化是利用波峰波谷比較法消除雜訊,取前後連續三點並以中間點比對左右點的差異度,可使圖片看起來比較平和不會充斥過多的白點。經由機台影像讀取後的圖片結果證明此改善想法可有效減少雜訊並強化影像,應用在多槽式鈔票整理機上的影像辨識達到即時快速改善影像。

並列摘要


The research was designed to develop an enhanced multi-slot banknote sorter and which features included recognition banknote true and false , banknotes counting , banknote image feature recognition and human machine interface (HMI) based on QT for Linux. The overall system consists of a control circuit of motor speed, the hardware circuit to identify the thickness of banknotes, the information of image entrance and the enhanced interface, the main control circuit board of advanced RISC machine-11 (ARM-11) hardware planning, and other related peripheral circuits. The system first reads the banknote of analog signal from the contact image sensor (CIS) and then amplifies the signal by the improved OP amplifier and converted the analog signal to eight bit digital grayscale of the image through a fast analog-digital converter (ADC). Furthermore, through the complex programmable logic device (CPLD) MAXII which used for image enhancement processing, the image would be reduced noise interference thus it would increase the image recognition level. First of all, the image signal by the amplifier and the rate adjusted to improve image contrast enhanced using a CPLD. Then, the method of image enhancement applies peak-valley compare method to decrease white spots. The method takes each three continuous point in the image and then compares the brightness of the second point with the first and third points. Through this way, it can avoid too many white points to be filled in the image information. From the experimental results, the images captured from the machine have a better contrast and it can effectively reduce the noises. After this, the enhanced images are suitable for the process we need. The result has been applied to the multi-slot banknote sorter to achieve an improved image.

參考文獻


[4] Piotr S. Windyga, “Fast Impulsive Noise Removal,” IEEE Trans. Image Processing, vol. 10, no. 1, 2001, pp.173-179.
[5] D. Brownrigg, “The Weighted Mesian filter,” Communication Association Computer, 1984, pp. 807-818.
[6] Deng, G. Cahill, L.W., “An adaptive Gaussian Filter for Noise Reduction and Edge Detection,” IEEE Conf. Nuclear Science Symposium and Medical Imaging. vol.3, 1993, pp. 1615-1619.
[7] SAMSUNG, S3C6410X USER'S MANUAL REV 1.10, SAMSUNG electronics, Aug. 2008.
[12] Fabrizio Russo, “An Image Enhancement Technique Combining Sharpening and Noise Reduction,” IEEE Trans. Instrumentation and Measurement, vol. 51, no. 4, 2002, pp. 824-828.

延伸閱讀