scholarly journals Verifikasi Citra Tanda Tangan Menggunakan Metode Prewitt dan Learning Vector Quantization

2018 ◽  
Vol 5 (2) ◽  
pp. 202-210
Author(s):  
Asfanji Sefta ◽  
Syarif Hidayatulloh

AbstrakTanda tangan adalah salah satu bukti persetujuan dari seseorang, Jadi tanda tangan ini memiliki arti yang sangat penting. Sering terjadi Kasus pemalsuan tanda tangan, antara lain disebabkan oleh sistem verifikasi yang tidak baik. Verifikasi tanda tangan ini kebanyakan dilakukan secara manual, Yaitu dengan membandingkan langsung dengan menggunakan mata Manusia yang memiliki banyak kelemahan. Jadi ketelitian dan keakuratan hasil yang diinginkan sering kurang memuaskan. Metode yang saya gunakan dalam membangun aplikasi verifikasi tanda tangan ini adalah dengan menggunakan metode Edge Detection dan metode Vector Quantization Learning. Program ini dibangun menggunakan Matlab. Berdasarkan hasil penelitian yang diperoleh dari pengujian sistem verifikasi tanda tangan dengan menerapkan metode edge detection operator prewitt dan metode Learning vector Quantization dengan penghitung vektor, maka diperoleh kesimpulan bahwa Metode Edge Detection operator prewitt dan Learning Vector Quantization dengan penghitung vektor dapat mengekstraksi fitur tanda tangan untuk memproses vektor yang digunakan dalam penghitungan vektor untuk mengenali tanda tangan yang asli dan yang palsu pada aplikasi verifikasi tanda tangan yang  membantu memverifikasi tanda tangan sehingga meminimalisasi pemalsuan tanda tangan.  Kata Kunci : Citra, Edge Detection, Vector Quantization Learning, Tanda Tangan AbstractThe signature is one of the proof of approval from a person, so this signature has a very important meaning. There are often cases of signature forgery, partly due to a poor verification system. This signature verification is mostly done manually, that is by comparing it directly using the eyes of a human who has many weaknesses. So the accuracy and accuracy of the desired results are often unsatisfactory. The method that I use in building this signature verification application is to use the Edge Detection method and the Vector Quantization Learning method. This program is built using Matlab. Based on the research results obtained from the signature verification system testing by applying edge detection operator prewitt method and Learning vector Quantization method with vector counters, the conclusion is that the Edge Detection Method operator prewitt and Learning Vector Quantization with vector counters can extract signature features to process vectors used in vector counting to identify original and fake signatures in the signature verification application that helps verify the signature so as to minimize signature falsification. Keywords: Image, Edge Detection, Vector Quantization Learning, Signature

2018 ◽  
Vol 5 (2) ◽  
pp. 202-210
Author(s):  
Asfanji Sefta ◽  
Syarif Hidayatulloh

AbstrakTanda tangan adalah salah satu bukti persetujuan dari seseorang, Jadi tanda tangan ini memiliki arti yang sangat penting. Sering terjadi Kasus pemalsuan tanda tangan, antara lain disebabkan oleh sistem verifikasi yang tidak baik. Verifikasi tanda tangan ini kebanyakan dilakukan secara manual, Yaitu dengan membandingkan langsung dengan menggunakan mata Manusia yang memiliki banyak kelemahan. Jadi ketelitian dan keakuratan hasil yang diinginkan sering kurang memuaskan. Metode yang saya gunakan dalam membangun aplikasi verifikasi tanda tangan ini adalah dengan menggunakan metode Edge Detection dan metode Vector Quantization Learning. Program ini dibangun menggunakan Matlab. Berdasarkan hasil penelitian yang diperoleh dari pengujian sistem verifikasi tanda tangan dengan menerapkan metode edge detection operator prewitt dan metode Learning vector Quantization dengan penghitung vektor, maka diperoleh kesimpulan bahwa Metode Edge Detection operator prewitt dan Learning Vector Quantization dengan penghitung vektor dapat mengekstraksi fitur tanda tangan untuk memproses vektor yang digunakan dalam penghitungan vektor untuk mengenali tanda tangan yang asli dan yang palsu pada aplikasi verifikasi tanda tangan yang  membantu memverifikasi tanda tangan sehingga meminimalisasi pemalsuan tanda tangan.  Kata Kunci : Citra, Edge Detection, Vector Quantization Learning, Tanda Tangan AbstractThe signature is one of the proof of approval from a person, so this signature has a very important meaning. There are often cases of signature forgery, partly due to a poor verification system. This signature verification is mostly done manually, that is by comparing it directly using the eyes of a human who has many weaknesses. So the accuracy and accuracy of the desired results are often unsatisfactory. The method that I use in building this signature verification application is to use the Edge Detection method and the Vector Quantization Learning method. This program is built using Matlab. Based on the research results obtained from the signature verification system testing by applying edge detection operator prewitt method and Learning vector Quantization method with vector counters, the conclusion is that the Edge Detection Method operator prewitt and Learning Vector Quantization with vector counters can extract signature features to process vectors used in vector counting to identify original and fake signatures in the signature verification application that helps verify the signature so as to minimize signature falsification. Keywords: Image, Edge Detection, Vector Quantization Learning, Signature


2014 ◽  
Vol 539 ◽  
pp. 141-145
Author(s):  
Shui Li Zhang

This paper presents new theorems Stevens edge detection method based on cognitive psychology on. Firstly, based on the number of the image is decomposed into high-frequency and low-frequency information, and the high-frequency information extracted by subtracting the maximum number of images to the image after the filter, then the amount of high frequency information into psychological cognitive psychology based on Stevenss theorem. The algorithm suppression refined edge after the non-minimum, applications Pillar K-means algorithm to extract image edge. Experimental results show that: the brightness of the image is converted to the amount of psychological edge can better unify under different brightness values.


2014 ◽  
Vol 511-512 ◽  
pp. 550-553 ◽  
Author(s):  
Jian Yong Liang

Edge detection is an old and hot topic in image processing, pattern recognition and computer vision. Numerous edge detection approaches have been proposed to gray images. It is difficult to extend these approaches to color image edge detection. A novel edge detection method based on mathematical morphology for color images is proposed in this paper. The proposed approach firstly compute vector gradient based on morphological gradient operators, and then compute the optimal gradient according to structure elements with different size. Finally, we use a threshold to binary the gradient images and then obtain the edge images. Experimental results show that the proposed approach has advantages of suppressing noise and preserving edge details and it is not sensitive to noise pixel. The finally edge images via the proposed method have high PSNR and NC compared with the traditional approaches.


Sign in / Sign up

Export Citation Format

Share Document