Visual detection of blemishes in potatoes using minimalist boosted classifiers

Barnes, Michael and Duckett, Tom and Cielniak, Grzegorz and Stroud, Graeme and Harper, Glyn (2010) Visual detection of blemishes in potatoes using minimalist boosted classifiers. Journal of Food Engineering, 98 (3). pp. 339-346. ISSN 0260-8774

This is the latest version of this item.

Documents
barnes09visualdetection.pdf
Preprint
[img]
[Download]
[img]
Preview
PDF
barnes09visualdetection.pdf - Whole Document

4MB

Full text URL: http://dx.doi.org/10.1016/j.jfoodeng.2010.01.010

Abstract

This paper introduces novel methods for detecting blemishes in potatoes using machine vision. After segmentation of the potato from the background, a pixel-wise classifier is trained to detect blemishes using features extracted from the image.
A very large set of candidate features, based on statistical information relating to the colour and texture of the region surrounding a given pixel, is first extracted.
Then an adaptive boosting algorithm (AdaBoost) is used to automatically select the best features for discriminating between blemishes and non-blemishes.
With this approach, different features can be selected for different potato varieties, while also handling the natural variation in fresh produce due to different seasons, lighting conditions, etc.
The results show that the method is able to build ``minimalist'' classifiers that optimise detection performance at low computational cost.
In experiments, blemish detectors were trained for both white and red potato varieties, achieving 89.6\% and 89.5\% accuracy, respectively.

Item Type:Article
Additional Information:This paper introduces novel methods for detecting blemishes in potatoes using machine vision. After segmentation of the potato from the background, a pixel-wise classifier is trained to detect blemishes using features extracted from the image. A very large set of candidate features, based on statistical information relating to the colour and texture of the region surrounding a given pixel, is first extracted. Then an adaptive boosting algorithm (AdaBoost) is used to automatically select the best features for discriminating between blemishes and non-blemishes. With this approach, different features can be selected for different potato varieties, while also handling the natural variation in fresh produce due to different seasons, lighting conditions, etc. The results show that the method is able to build ``minimalist'' classifiers that optimise detection performance at low computational cost. In experiments, blemish detectors were trained for both white and red potato varieties, achieving 89.6\% and 89.5\% accuracy, respectively.
Keywords:Potatoes, Machine Vision, quality control, AdaBoost, Adaptive boosting, Matlab
Subjects:G Mathematical and Computer Sciences > G400 Computer Science
D Veterinary Sciences, Agriculture and related subjects > D610 Food Science
G Mathematical and Computer Sciences > G740 Computer Vision
Divisions:College of Science > School of Computer Science
ID Code:2206
Deposited By: Michael Barnes
Deposited On:21 Jul 2010 13:20
Last Modified:18 Nov 2013 16:09

Available Versions of this Item

  • Visual detection of blemishes in potatoes using minimalist boosted classifiers. (deposited 21 Jul 2010 13:20) [Currently Displayed]

Repository Staff Only: item control page