AE—Automation and Emerging Technologies: Artificial Intelligence Classifiers for sorting Apples based on Watercore
✍ Scribed by M.A. Shahin; E.W. Tollner; R.W. McClendon
- Publisher
- Elsevier Science
- Year
- 2001
- Tongue
- English
- Weight
- 302 KB
- Volume
- 79
- Category
- Article
- ISSN
- 0021-8634
No coin nor oath required. For personal study only.
✦ Synopsis
Statistical classi"cation methods, such as the Bayesian classi"er, can provide optimal classi"cation but their performance depends heavily on the assumption of normality of the input data. Arti"cial intelligence (AI) approaches, on the other hand, entail less stringent assumptions about the statistical characteristics of the input data. Hence, the neural network and fuzzy logic classi"ers are expected to perform better than the Bayesian classi"er for a given data set. This paper describes steps involved in the development of an optimal neural network classi"er and a fuzzy classi"er for sorting apples using the selected image features as the input variables. Performance of the AI classi"ers developed was compared with that of the Bayesian classi"er using the same data set. The fuzzy classi"er (80%) performed as well as the Bayesian classi"er with linear discriminant functions (79%), whereas, the neural classi"er performed better (88%).