Size estimation of sweet onions using consumer-grade RGB-depth sensor

A multimodal machine vision system for quality inspection of onions

“…The rationales for selecting these image techniques were: color imaging provides spatial and color information of onions; spectral imaging can be used to examine surface blemish and rot of onions; RGB-D sensor can measure size and geometry information of onions based on the topological data it collects; X-ray imaging can be used to detect internal defects and disease infections in onions. The potential of selected technologies on onion quality evaluation have been proven by previous studies (Tollner et al, 2005;Wang and Li, 2014;Wang et al, 2012a). This work focused on the integration of these techniques to make a comprehensive evaluation of the quality of onions.…”

“…Based on the measured weight and volume, the density of onion was calculated. The detailed algorithms for calculating these traits were described in Wang and Li (2014). The estimated diameter, volume, weight, and density of onions were compared to their corresponding ground truth values.…”

“…Wang et al (2013) used hyperspectral imaging in 400-1000 nm to quantitatively predict the soluble solids content and dry matter content of onions. Wang and Li (2014) proposed a RGB-D method to measure the size traits of onions based on their point cloud images. Each of these studies evaluated certain quality traits of onions.…”

Advances in Structured Light Sensors Applications in Precision Agriculture and Livestock Farming

“…With reference to first processing agrifood industry, Wang and Li (2014) propose a new Kinect-based technique to measure the diameter and volume of sweet onions. Becoming robotization a reality in the slaughtering process, Kinect sensor could be a cheap and feasible alternative for recognizing meat cuts according to Larsen et al (2014).…”

Contribuciones al desarrollo de un sistema efectivo para la inspección de cultivos

“…Los resultados sugieren que este sensor puede ser un dispositivo de alta precisión para estimar el volumen de las malas hierbas y determinar el estado del cultivo. En (Wang & Li, 2014) se usa un sensor Kinect v2 para estimar el volumen de cebollas, mostrando que el volumen calculado estaba directamente relacionado con el estimado con las medidas procedentes del sensor.…”