|Tao, Yang - UNIV OF MD|
|Cheng, Mei - UNIV OF MD|
Submitted to: Acta Horticulture Proceedings
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 14, 2002
Publication Date: N/A
Interpretive Summary: Automation of apple defect sorting is highly demanded by the apple industry, since apple sorting via a machine vision system can significantly improve process efficiency, reduce labor cost, and improve sorting accuracy. Although there have been intensive research studies in this field, the development of an on-line sorting system for apples has been unsuccessful primarily due to the presence of apple stem-end/calyx which can be falsely recognized as apple defects and sorted out. Therefore, developing a technology that can distinguish the truce defect from the stem-end/calyx of apples is essential to the success of this application. This research paper describes the development of a new technology that enables the machine vision system to distinguish the true defect from the stem-end /calyx of apples. By using a dual-camera vision system, images of apples illuminated under the lights of different wavelength were captured, allowing the identification of the defect and stem-end/calyx of apples. In addition, this paper also describes the development of algorithms that needed to analyze and separate the defect from the stem-end/calyx of apples. Since defect sorting is an important postharvest process, the application of this technology will improve the efficiency of postharvest handling of apple industry and ensure the quality of apple products being supplied to consumers. In addition to apples, the methodology and algorithms developed in this research are also applicable to defect sorting and online inspection of other fruits and vegetables. The end results of this research will benefit processors and packers of apples, the fresh produce industry, as well as American consumers of fresh produce.
Technical Abstract: One of the difficult problems in automated machine vision apple sorting is the distinction between true defects and the stem-end/calyx. To solve this problem, a dual imaging approach using near infrared (NIR) and mid-infrared (MIR) was developed for combined sensing. Based on this method, the MIR is sensitive to the stem-end/calyx, whereas the NIR is sensitive to both stem-end/calyx and defects. Through image processing of combined images, the distinction is achieved. Experiments show the robustness of the method for on-line defect sorting of apples.