Image analysis and statistical modelling for measurement and quality assessment of ornamental horticulture crops in glasshouses

• Published in: Biosystems engineering 2009-10, Vol.104 (2), p.161-168
• Main Authors: Parsons, N.R., Edmondson, R.N., Song, Y.
• Format: Article
• Language: English
• Subjects: Agricultural machinery and engineering; Agronomy. Soil science and plant productions; bedding plants; Biological and medical sciences; crop quality; digital images; Fundamental and applied biological sciences. Psychology; Generalities. Biometrics, experimentation. Remote sensing; greenhouse production; greenhouses; image analysis; measurement; neural networks; ornamental plants; plant morphology; product grading; quality control; statistical models
• ISSN: 1537-5110; 1537-5129
• DOI: 10.1016/j.biosystemseng.2009.06.015

Image analysis for ornamental crops is discussed with examples from the bedding plant industry. Feed-forward artificial neural networks are used to segment top and side view images of three contrasting species of bedding plants. The segmented images provide objective measurements of leaf and flower cover, colour, uniformity and leaf canopy height. On each imaging occasion, each pack was scored for quality by an assessor panel and it is shown that image analysis can explain 88.5%, 81.7% and 70.4% of the panel quality scores for the three species, respectively. Stereoscopy for crop height and uniformity is outlined briefly. The methods discussed here could be used for crop grading at marketing or for monitoring and assessment of growing crops within a glasshouse during all stages of production.