In ovo sexing of eggs from brown breeds with a gender-specific color using visible-near-infrared spectroscopy: effect of incubation day and measurement configuration.

The culling of day-old male chicks is an animal welfare issue within the laying hen industry that raises substantial ethical concern. Alternative methods are sought to pre-select males during embryonic development. This method is called in ovo sexing and allows more humane male culling. A robust and non-invasive in ovo color sexing technique was developed and validated in this research. To this end, visible-near-infrared (vis-NIR) point spectroscopy was used, which has advantages over state-of-the-art hyperspectral imaging in terms of accuracy and cost. Two independent experiments were each conducted on a batch of 600 Isa Brown eggs. These eggs were individually illuminated on d 8 to 14, and d 18 of incubation by a halogen lamp and the signal was measured in the vis-NIR range from 300 to 1,145 nm. Next, optimal preprocessing strategies were applied and partial least squares discriminant analysis (PLS-DA) models were built and further optimized after performing a forward interval partial least squares (FiPLS) for variable selection. Results demonstrated that d 12 is too early for vis-NIR in ovo sexing, resulting in a prediction accuracy of 86.49%. However, after 13 d of incubation, an accuracy of 97.78% was obtained, increasing to 99.52% on d 14. Furthermore, these accuracies were higher than earlier reported percentages from hyperspectral imaging and successful sexing was expedited from d 14 to d 13. Moreover, prediction improvement up to 99.05% was obtained on d 13 by correcting for the variability in eggshell properties using d 0 eggshell corrections. Applying the method on d 18 resulted in a lower accuracy of 94.62% due to stronger light attenuation by the growing embryos. Finally, a reduced spectral range of 749 to 861 nm was found to be sufficient for correct classification of 98.46% of the eggs. This paves the way for high-throughput and cost-efficient usage of smaller and cheaper spectrophotometers in commercial hatcheries.

Augmented design and analysis of computer experiments: a novel tolerance embedded global optimization approach applied to SWIR hyperspectral illumination design.

A novel meta-heuristic approach for minimizing nonlinear constrained problems is proposed, which offers tolerance information during the search for the global optimum. The method is based on the concept of design and analysis of computer experiments combined with a novel two phase design augmentation (DACEDA), which models the entire merit space using a Gaussian process, with iteratively increased resolution around the optimum. The algorithm is introduced through a series of cases studies with increasing complexity for optimizing uniformity of a short-wave infrared (SWIR) hyperspectral imaging (HSI) illumination system (IS). The method is first demonstrated for a two-dimensional problem consisting of the positioning of analytical isotropic point sources. The method is further applied to two-dimensional (2D) and five-dimensional (5D) SWIR HSI IS versions using close- and far-field measured source models applied within the non-sequential ray-tracing software FRED, including inherent stochastic noise. The proposed method is compared to other heuristic approaches such as simplex and simulated annealing (SA). It is shown that DACEDA converges towards a minimum with 1 % improvement compared to simplex and SA, and more importantly requiring only half the number of simulations. Finally, a concurrent tolerance analysis is done within DACEDA for to the five-dimensional case such that further simulations are not required.