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.

Characterization and Identification of Cracks on Perforated Embryo Eggs Based on Fractal Theory

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:  Applied Engineering in Agriculture. 40(3): 317-326. (doi: 10.13031/aea.15902) @2024
Authors:   Wenjie Lin, Mingyan Zhao, Haoke Chen, Yuhan Fang, Kaixiang Zhang
Keywords:   Characterization analysis, Crack, Fractal theory, Image processing, Linear discrimination.

Highlights

A fractal matrix model was established to characterize the cracks on the surfaces of perforated embryo eggs.

A function for classifying perforated embryo eggs was constructed using the Fisher linear discriminant algorithm.

An experiment to identify the qualification of perforated embryo eggs was performed.

Abstract. In response to the problems of low accuracy and unsatisfactory consistency in manual identification, a method for characterizing and identifying cracks on perforated embryo eggs is proposed based on fractal theory. Due to the intersection of the fractal dimensions of a few qualified and cracked embryo eggs, an improved fractal box-counting algorithm was used to characterize the overall characteristics of the cracks and construct an 8 x 8 fractal matrix using image segmentation technology. Based on the morphology and structure, the eggshell was divided into three regions and five characteristic variables expressing perforation cracks were extracted. Two significant variables were screened using the stepwise method to construct a linear discriminant function. Ten-day-old embryonated eggs of white phoexix black-bone chickens were selected as experimental materials. Embryo egg perforation and image acquisition devices were established. Then, crack recognition experiments were carried out. The accuracy of crack determination was 98.89%, and the identification time for a single egg was less than 1.24 s. The research results can lay a solid theoretical and practical foundation for the development of automatic recognition systems for perforated embryo eggs.

(Download PDF)    (Export to EndNotes)