Table of Content

Open Access iconOpen Access

ARTICLE

The Crime Scene Tools Identification Algorithm Based on GVF‐Harris‐SIFT and KNN

Nan Pan1, Dilin Pan2, Yi Liu2

1 Faculty of Civil Aviation and Aeronautical, Kunming University of Science & Technology, Kunming 650500, P.R. China
2 Kunming SNLab Tech Co., Ltd., Kunming 650228, P.R. China

* Corresponding Author: Nan Pan, email

Intelligent Automation & Soft Computing 2019, 25(2), 413-419. https://doi.org/10.31209/2019.100000103

Abstract

In order to solve the cutting tools classification problem, a crime tool identification algorithm based on GVF-Harris-SIFT and KNN is put forward. The proposed algorithm uses a gradient vector to smooth the gradient field of the image, and then uses the Harris angle detection algorithm to detect the tool angle. After that, the descriptors of the eigenvectors in corresponding feature points were using SIFT to obtained. Finally, the KNN machine learning algorithms is employed to for classification and recognition. The experimental results of the comparison of the cutting tools show the accuracy and reliability of the algorithm.

Keywords


Cite This Article

APA Style
Pan, N., Pan, D., Liu, Y. (2019). The crime scene tools identification algorithm based on gvf‐harris‐sift and KNN. Intelligent Automation & Soft Computing, 25(2), 413-419. https://doi.org/10.31209/2019.100000103
Vancouver Style
Pan N, Pan D, Liu Y. The crime scene tools identification algorithm based on gvf‐harris‐sift and KNN. Intell Automat Soft Comput . 2019;25(2):413-419 https://doi.org/10.31209/2019.100000103
IEEE Style
N. Pan, D. Pan, and Y. Liu "The Crime Scene Tools Identification Algorithm Based on GVF‐Harris‐SIFT and KNN," Intell. Automat. Soft Comput. , vol. 25, no. 2, pp. 413-419. 2019. https://doi.org/10.31209/2019.100000103



cc This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 1333

    View

  • 976

    Download

  • 0

    Like

Share Link