GA-Based Optimization of SURF Algorithm and Realization Based on Vivado-HLS

Creative Commons License

Ozdemir H., Sever R., POLAT Ö.

TRAITEMENT DU SIGNAL, cilt.36, sa.5, ss.377-382, 2019 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 36 Sayı: 5
  • Basım Tarihi: 2019
  • Doi Numarası: 10.18280/ts.360501
  • Dergi Adı: TRAITEMENT DU SIGNAL
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.377-382
  • Anahtar Kelimeler: speeded-up robust features, high-level synthesis, genetic algorithm, optimization, character recognition
  • Akdeniz Üniversitesi Adresli: Evet

Özet

The aim of this study is realization of SURF algorithm based on Vivado-HLS tool for FPGA platform. The SURF algorithm is a method which is used in image processing that is not affected by feature changes such as size, color and contrast. Genetic algorithm is used to determine the optimum values of the parameters affecting the success of the SURF algorithm in this work. It has been observed that when the parameter values determined using the optimization algorithm is used, the success rate significantly increases. The proposed method has been tested for character recognition application with fixed font which is independent from rotation and character size. In this work, the tests were carried out with images formed from numbers. Reference and test pictures for each number were created. The test images consist of images of different sizes and rotations. Optimal parameter values were used in HLS after being determined by proposed approach. Using the proposed optimized SURF structure, high success rates were obtained.