29 May 2013 A noninvasive technique for real-time detection of bruises in apple surface based on machine vision
Author Affiliations +
Abstract
Apple is one of the highly consumed fruit item in daily life. However, due to its high damage potential and massive influence on taste and export, the quality of apple has to be detected before it reaches the consumer’s hand. This study was aimed to develop a hardware and software unit for real-time detection of apple bruises based on machine vision technology. The hardware unit consisted of a light shield installed two monochrome cameras at different angles, LED light source to illuminate the sample, and sensors at the entrance of box to signal the positioning of sample. Graphical Users Interface (GUI) was developed in VS2010 platform to control the overall hardware and display the image processing result. The hardware-software system was developed to acquire the images of 3 samples from each camera and display the image processing result in real time basis. An image processing algorithm was developed in Opencv and C++ platform. The software is able to control the hardware system to classify the apple into two grades based on presence/absence of surface bruises with the size of 5mm. The experimental result is promising and the system with further modification can be applicable for industrial production in near future.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Juan Zhao, Juan Zhao, Yankun Peng, Yankun Peng, Sagar Dhakal, Sagar Dhakal, Leilei Zhang, Leilei Zhang, Akira Sasao, Akira Sasao, } "A noninvasive technique for real-time detection of bruises in apple surface based on machine vision", Proc. SPIE 8721, Sensing for Agriculture and Food Quality and Safety V, 87210O (29 May 2013); doi: 10.1117/12.2015897; https://doi.org/10.1117/12.2015897
PROCEEDINGS
8 PAGES


SHARE
RELATED CONTENT

Corn plant locating by image processing
Proceedings of SPIE (February 01 1991)
Image processing to locate corn plants
Proceedings of SPIE (March 01 1991)

Back to Top