Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Abstract. Early bruises detection of apples with different colors and textures on their surfaces is of great importance yet challenging by using traditional diagnosis and hyper imaging technologies. This paper proposed a novel method by using a multi-resolution microwave imaging (MMI) technology that is tailored for apple‘s early bruises detection. High accuracy and efficiency can be optimally achieved since MMI is an effective yet fast non-invasive technique that is sensitive to the dielectric constants and material properties of fruit, hence make it suitable for water-content characterization in agricultural and food industry. In the proposed MMI approach, firstly, the wide-band microwave sensing data of apples with and without early bruises were acquired using tailored GHz transducer. Then, the characteristic frequencies were extracted as fingerprints of apples by using pattern classification algorithms such as Principal Component Analysis (PCA). Finally, a prediction model for apple‘s early bruises damage detection was developed and the feasibility of the proposed system could be validated by using a large amount of apple samples with different varieties and contamination levels.