Boosting minimalist classifiers for blemish detection in potatoes

Barnes, Michael and Duckett, Tom and Cielniak, Grzegorz (2009) Boosting minimalist classifiers for blemish detection in potatoes. In: Image and Vision Computing New Zealand, 23-25 November 2009, Wellington, New Zealand.

Documents
Boosting minimalist classifiers for blemish detection in potatoes
[img]
[Download]
[img] PDF
Boosting_Minimalist_Classifiers_for_Blemish_Detection_in_Potatoes.pdf

1MB

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 nonblemishes.
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, minimalist blemish detectors were trained for both white and red potato varieties, achieving 89.6% and 89.5% accuracy respectively.

Item Type:Conference or Workshop Item (Paper)
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 nonblemishes. 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, minimalist blemish detectors were trained for both white and red potato varieties, achieving 89.6% and 89.5% accuracy respectively.
Keywords:Machine Learning, AdaBoost, Potatoes, Quality Control, Classification, Pixelwise
Subjects: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:2134
Deposited By: Michael Barnes
Deposited On:15 Jan 2010 08:52
Last Modified:18 Jul 2011 16:22

Repository Staff Only: item control page