INTERNATIONAL JOURNAL FOR RESEARCH & DEVELOPMENT IN TECHNOLOGY Volume-5,Issue-5 (May-16) ISSN (O) :- 2349-3585 LABVIEW DESIGN FOR EDGE DETECTION USING LOG GABOR FILTER FOR DISEASE DETECTION Vipul Kumbhalwar 1, Swati Dixit 2 12 M.Tech student, Department of Electronics and Telecommunication, 12 M.Tech student, G.H.Raisoni College Of Engineering and Technology, Maharashtra, India. Abstract Edge detection is an important tool in the field of image processing, Disease like tonsillitis, tumor, fracture and many more can be detect and cured in its early stage, by detecting the edges of that disease, so edge detection having always given first attention in the field of image processing. In this paper tonsillitis detection module is design in LabVIEW and then it implemented on NI Sbrio 9631 FPGA kit. The algorithm used is SOBEL operator and for best ridges and fast processing log gabor filter is used. It is used in various applications as medical image processing, object detection etc. The main aim behind this is to process the image and use it in various applications using FPGA platform. Field Programable Gate Array(FPGA) has an huge embedded multipliers as well as large amount of internal memory for real time application which is use in digital image processing, by this way parallisiom is possible. Hence, Field Programmable Gate Array always provides the platform for real time image processing with higher performance as compare to microprocessor and DSPs (Digital Signal Processors). The FPGA image preprocessing system architecture which uses Sobel algorithm and log gabor filter to realize the edge detection is proposed here. LabView NI Vision assistant is used for better performance and for simplified design, by using NI Vision Assistant different image pre-processing can be done. Index Terms- Edge detection, Sobel operator, Log-Gabor filter, LabVIEW 14.0, NI Vision Assistant, LabVIEW FPGA. I-INTRODUCTION Image processing is always having the important, broad, fundamental and active area in the field of medical, surveillance, authentication and many more application. In term of medical disease, accurate results is always preferred, Medical applications always consist with the different types of image processing techniques like image enhancement method, image pre-processing image post-processing, focused area selection and object detection etc, and this all techniques/methods depend on the edge detection, hence edge detection is always sensitive and research area. Detection of some disease like tonsillitis, tumor and fracture is depend on detection of edges of disease and for this, pixel to pixel calculation is performed in image processing. In this paper one of the finest edge detection method is used that is sobel edge operator, mask of sobel operator is work on the entire image and give us sharp and accurate edges so that disease can be detect, and the algorithm is implemented on FPGA kit. The main aim behind this procedure is to process the image processing and use it in various applications for medical application using FPGA. LabVIEW software is very active and powerful tool, When it comes to creating DAQ applications. LabVIEW includes a set of Virtual instruments ( VI ), that it let you configure, data acquire from, and transfer data to data acquisition (DAQ) devices. Often, in LabVIEW one device could perform a variety of functions, such as analog to digital (A/D) conversion, digital to analog (D/A) conversion, digital input and output, as well as counter/timer operation, Each device supports different DAQ and signal generation speedswith respect to image processing. Also, each data acquisition device is designed for specific hardware, platforms and operating systems for digital image processing. After reading the pixels of an image, the algorithm is applied in VHDL, then processing the image on FPGA as an hardware implementation, edge detected image is displayed on LabVIEW front panel. The entire simulation of the above process is done VHDL using XILINX NI Sbrio 9631 FPGA, and to display input and output image Lab VIEW is used. Field Programmable Gate Array (FPGA) is a reconfigurable device and because of use of such devices the time to market cost and time reduces. Also it becomes very easy 56 All rights reserved by www.ijrdt.org
for the result verification and debugging processing. FPGA implementation could become easy because of NI Vision Assistant as it having scripts format which can easily implemented on any image. Design scripts can covert for LabVIEW code which is possible for edge detection. In Labview image processing image acquisition, image processing, different types of filter can use for noise remove per pose. With the help of Math tool scripts kit we can call any mat lab function to LabVIEW as this is the advantage for LabVIEW user. Here Log Gabor filter is called from matlab for noise remove purpose and for fast processing. Fig 2. Image Acquition in LabVIEW II.PROPOSED METHODS A. Block Diagram for Disease Detection algoritham After image acquisition part LabVIEW apply some basic Block diagram of disease detection module is shown below, it processing for fine result and for better noise reduction so that having different blocksets which used for image processing resultant output will be accurate. perpose. In LabVIEW for edge detection colo plane C. Image processing by Log Gabor filtering extraction, image thresholding, IMAQ mathtool kit is Basically Log Gabor filter is used for image filtering and for best ridges of an image, which is shown in system. Fig 1. Proposed disease detection model B. Image Acquisition with IMAQ and Preprocessing In given design camera is used for real time application, camera will capture image that will be normal image or disease detected image and later it transfer to LabVIEW for image processing. IMAQ tool detect input image, acquire it for LabVIEW processing. In image processing first RGB image converted in grayscale image with the help of color plane extraction. Color plane extraction extract one basic color from RGB image and at output grayscale image we found, Here mathlookup (exponential) and image thresholding function used for background color adjustment. Fig 3. Implementation of log gabor filter D. Sobel Edge Operator Sobel edge operator is nothing but a edge detecting methodology used for finding edge detection in digital image processing. Result of sobel edge detector is quite better as compare to other edge detector like canny edge detector, Robert operator, Prewit operator and laplasian filter. Mask of sobel edge detector having two filter Hx filter and Hy filter, one for horizanal pixel operation and other one is for vertical pixel 57
operation which mean horizontal edge detection and vertical preprocessing, Log gabor filtering, mathlookup and image edge detection. thresholding. following is the sample matrix for vertical and horizontal edge Block diagram of disease detection module represent different detection of sobel operator. type of processes used for finding the disease detection. Block diagram process all parameter and then results show on front panel. Here the direction of edges can be determine by using following formula, GM(x,y)= 2 Hx Hy 2 Fig 4. Sobel edge detection by LabVIEW VI Fig 5. Result of sobel edge detector III. LABVIEW CODING FOR DISEASE DETECTION Fig 6- Block diagram of Disease detection module Here LabVIEW coding for tonsillitis disease detection is shown below, which consist of an image acquisition, image 58
IV. HARDWARE IMPLEMENTATION In Digital image processing where the hardware and Software both combination comes, the testing get reduced and V. PROJECT EXPLORER The project explorer window shows different parts that a Labview project constitutes. An addition of VI s can be done performance will increase, as software and hardware both here. The project explorer shows the FPGA target which create the strongest parameter in medical application. As only software has become less meaningful as image size and bit specifies the FPGA board, real time VI or also called as host VI and the FPGA VI.The FPGA target runs on FPGA VI. depths grows larger. FPGA are used for high speed processing in images. With the development of FPGA, a large amount of data are captured using satellite and ground based detection systems. Here in image processing, Labview platform consist of NI Single Board RIO 9631 ( SbRIO 9631). Single board rio is a product from national instruments which has XILINX Spartan 3 FPGA in it. It also consist of a microprocessor which is from Freescale Semiconductor.It also has analog I/O and digital I/O. Fig.9(a), Project Explorer VI. VHDL Bit file Genaration Below figure showing VHDL bit file genaration for hardware implementation, Fig 7. NI Sbrio FPGA kit Implementation of disease detection module possible only when the block diagram design will only in LabVIEW fpga tool. For the implementation on LabVIEW FPGA, we required VI file and Host file. NI SbRIO is an high speed FPGA tool used in digital image processing..fig. 9 (b) VHDL Bit file Genaration 59 Fig 8. FPGA Target VI. Fig 9 (c). Bit File generation report
After the code generation successfully, VHDL bit file dump into SbRIO 9631 for the hardware implementation.when result compare, found that processing time of software and hardware is different, as hardware took less time for operation as compare to software. Fig 10(c) Disease detection with sobel operator Fig 9 (d ). Code deployment process VII. RESULTS Result of disease detection by using Sobel operator, Log gabor filter and with LabVIEW are shown below in fig 10(a), (b),(c), and (d), Fig 10(c) Disease detection with sobel operator Fig 10(a) Disease detection with sobel operator Fig 10(b) Disease detection with sobel operator VIII. CONCLUSION Here the paper proposed the disease detection module with the help of sobel edge detection algorithm and log gabor filter. It conclude that the LabVIEW is totally compatible for digital image processing. Also paper proposed that the hardware and software architecture for the sobelcedge detection which is designed for the NI Single board RIO FPGA platform. As LabVIEW is graphical programming language it is easy to understand and also easy for implementation. VIII. REFERENCES [1]K. Sudharani, A.swapnarani, K.manikumari,t. C. Sarma. Satya prasad labview based brain tumor area and length detection in ct and mri scan images International Journal of Advanced Trends in Computer Science and Engineering (IJATCSE), Vol.2, No.5, Pages :70-74 (2013) Special Issue of ICCECT 2013 - Held during September 20, 2013, Bangalore, India. [2] Kumar A.V, Nataraj K.R Result Analysis of LabVIEW and MatLab in Application of Image Edge 60
Detection (International Journal of Computer Applications (0975 888) Volume 48 No.9, June 2012). [3] Hossein Ghayoumi Zadeh, Siamak Janianpour, and Javad Haddadnia Recognition and Classification of the Cancer For Technological Research In Engineering Volume 2, Issue 9, May-2015. [13] A.R.Bhagat, S.R. Dixit, A. Y. Deshmukkh Sobel Edge Detection for Real-Time Applications, International Journal Cells by Using Image Processing and LabVIEW For Technological Research In Engineering Volume 2, Issue (International Journal of Computer Theory and Engineering, Vol. 5, No. 1, February 2013). [4] Swati R. Dixit, Amol Y. Deshmukh Design Strategies For Various Edge Detection Techniques for Disease Detection International Journal of Scientific Engineering and Applied Science (IJSEAS) - Volume-1, Issue-8,November 2015. [5] V. Kamatchi Sundari1,M. Manikandan, P.Prakash FPGA 9, May-2015. [14] Meenal Bhatia,Swati.R.Dixit Edge detection for X-Ray image using Labview,International Journal of Advanced Research in Computer and Communication Engineering (IJARCCE ), Volume 5, Issue 3, (March 2016), e-issn: 2278 1021, p-issn No : 2319 5940. [15] J.Canny, A computational approach to edge detection, IMPLEMENTATION of SOBEL EDGE DETECTOR IEEE transaction on pattern analysis and machine International Journal of Advances in Science and Technology (IJAST). [6] Harshada Shimpi1, NishaGaikwad, Meghana Dhage, intelligence, vol 8, no.6, pp,679-698,nov 1986. [16] Ravi Kumar A.V, Dr. Nataraj K.R, Dr.Rekha K.R Morphological Real Time Video Edge Detection in NI Prof.A.S.Pawar Implementation of Edge Detection Labview International Journal of Computer Science and Algorithm Using FPGA International Journal of Advanced Information Technologies, Vol. 3 (2), 2012,3808-3811. Research in Electrical, Electronics and Instrumentation Engineering, Vol. 4, Issue 4, April 2015. [7] Dr. Abdulsattar M. Khidhir1, Nawal Younis Abdullah FPGA Based Edge Detection Using Modified Sobel FilterGA Based Edge Detection Using [8] Modified Sobel Filter, International Journal for Researchand Development in Engineering (IJRDE) Vol.2 Issue.1, June-July 2013. [9] Pradnya Jadhav, Usha Jadhav Sobel Edge Detection implementation using FPGA International Journal of Science and Research (IJSR). [10] Bharath K, Dr. N. G. Kurahatti, VERILOG DESIGN FOR FEATURE EXTRACTION USING LOG- GABOR FILTER FOR DISEASE DETECTION International Journal For Technological Research In Engineering Volume 2, Issue 9, May-2015. [11] Gautam Appasaheb Kudale, Mahesh D. Pawar Study and Analysis of Various Edge Detection Methods for X-Ray Images ( International Journal of Computer Science and Application Issue 2010. [12] A.R.Bhagat, S.R. Dixit, A. Y. Deshmukkh Sobel Edge Detection for Real-Time Applications, International Journal 61