The worldwide production of poultry meat was estimated to reach 116 million metric tons in 2016. About ten percent of all broiler-type eggs is not fertilized. After the breeding process, these unhatched eggs do not have commercial value and are destined to be destroyed. Moreover, valuable resources are wasted during incubation and the risk of contamination is increased. Detection of the fertilization of eggs in an early stage of the breeding process is beneficial for both environmental and commercial interest. This research aims to conceive a practical method to detect fertility of an egg in an early stage of breeding, preferably before incubation.
This research is divided in two stages, each with two different experiments based on spectroscopy. In the first stage, the usability of the high-frequency radio band for fertility detection is investigated. The second experiment of this stage concentrates on the usability of the visible-light/near-infrared band (400-1600 nm). Based on the results of the first stage, the second stage focuses on two new experiments. One experiment aims to verify a fertility detection method by measuring a large amount of eggs. Another experiment explores the feasibility of detecting rotten (“exploder”) eggs for increased accuracy in current fertility detection methods.
All findings are described in this report. Additionally, an accurate and flexible prototype measurement system is developed using industrial components. This system shows the feasibility of applying measurement methods in industrial environments. Further research topics are suggested that can use the findings in report to improve both detection stage and accuracy in practice.