The use of scale invariant feature transform (SIFT) algorithms to identification garbage images based on product label
Authors:
- 1 October 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 336-341
Abstract
Garbage has become a serious problem, especially in major cities. Garbage basically contain elements of organics and non-organic mixed. In a large volume, the separation of organic and non-organic become difficult jobs and longer if done conventionally. Separation is the identification of objects with one another to be classified. Along with the development of technology, it is possible to design intelligent machines capable of identifying organic and non-organic objects, then the machine can separate quickly. Attributes organic and non-organic element can be recognized from the product label and can be a special characteristic. In this experiment, the algorithm SIFT (Scale invariant Feature Transform) to extract the characteristics of the image garbage label. This technique makes use keypoint or image features garbage properly. Based on the experimental results, the method of identification of organic and non-organic using SIFT algorithm gives results with an average accuracy of 89.9%. With a high level of complexity of a large garbage collection, the performance results including good category.Keywords
This publication has 2 references indexed in Scilit:
- An Introduction to Object Recognition2010
- Distinctive Image Features from Scale-Invariant KeypointsInternational Journal of Computer Vision, 2004