Signify (Netherlands)

We present a systematic derivation of several mathematical models for inverse freeform optical design, based on Hamilton’s characteristic functions and conservation of luminous flux. Optimal transport theory is the correct mathematical framework to formulate optical design problems. We apply this theory to close our models, such that these are suitable for numerical simulation. We outline several iterative least-squares solvers and demonstrate their performance for a few example problems.

We present an inverse method for transforming a given parallel light emittance to two light distributions at different parallel target planes using two freeform reflectors. The reflectors control both the spatial and directional target coordinates of light rays. To determine the shape and position of the reflectors, we derive generating functions and use Jacobian equations to find the optical mappings to the two targets. The model is solved numerically by a three-stage least-squares algorithm. A feasibility condition is derived, which ensures that the reflectors are not self-intersecting. Several examples validate this condition and demonstrate the algorithm's capability of realizing complex distributions.

Natural Language Processing (NLP) has witnessed significant advancements, particularly with the development of multilingual models. However, a persistent challenge remains: the resource asymmetry among languages. This interdisciplinary analysis bridges the gap between computational linguistics, sociolinguistics, and ethics to diagnose resource asymmetry in multilingual NLP. We introduce metrics (RPI, LCS, TECS) quantifying disparities and demonstrate, through case studies, how policy and community-driven models can foster equitable language. This paper provides a comprehensive analysis of the current state of resource asymmetry in multilingual NLP, examining the contributing factors, evaluating existing approaches, and identifying research gaps. Also, ethical frameworks and community-driven strategies are proposed to bridge the gaps in NLP, emphasizing participatory design and equitable resource allocation.

We present an inverse method to design freeform reflectors that transform a parallel source with given emittance into desired illuminances at two target screens. Our model is based on Hamilton’s characteristic functions and energy conservation.

In this paper, we discuss a mathematical model for inverse freeform design of an optical system with two reflectors in which light transfers from a point source to a point target. In this model, the angular light intensity emitted from the point source and illuminance arriving at the point target are specified by distributions. To determine the optical mapping and the shape of the reflectors, we use the optical path length and take energy conservation into account, through which we obtain a generated Jacobian equation. We express the system in both spherical and stereographic coordinates, and solve it using a sophisticated least-squares algorithm. Several examples illustrate the algorithm's capabilities to tackle complicated light distributions.

Supplementary far-red light may increase fruit yield in greenhouse tomato cultivation. Apart from end-of-day applications, research has focussed on the effects of far-red light during the entire photoperiod, while few studies focused on the timing of far-red application. In this study, we assessed the impact of timing and duration of far-red application on fruit yield and quality in a high-wire tomato crop. Two commercial tomato cultivars were grown under 226 μmol m−2 s−1 of red and white supplementary LED lighting. Far-red light (63 μmol m−2 s−1) was additionally supplied during either the first or the second half of the photoperiod. These treatments were compared to a negative control, where no far-red light was supplied, and a positive control, where far-red light was supplied during the whole photoperiod. A yield component analysis was conducted to identify the parameters responsible for changes in fruit yield. We determined that fruit yield, total fruit dry weight and fruit production per unit of supplementary PAR light (400–700 nm) increased linearly with the duration of far-red application. Fruit yield was increased to a similar extent whether far-red was supplemented in the first or in the second half of the photoperiod. While different light treatments included 0, 14, or 28 % far-red in their cumulative supplementary photon flux density, there were no significant differences in their fruit production per unit of supplementary radiation (400–800 nm). Supplementary far-red increased fruit dry weight per unit of cultivation area by 23 % to 25 % as a result of increasing both plant dry weight (10 % to 12 %) and dry weight partitioning to the fruits (10 % to 14 %). The increase in fruit yield was less than the corresponding increase in total fruit dry weight due to a simultaneous increase in fruit dry matter content. Total soluble solids and pH of the fruits were hardly affected by far-red application. We concluded that supplementary far-red increased tomato fruit yield without compromising fruit quality, with this increase being determined by the duration of far-red application, not by its timing during the photoperiod.

We present several geometrical models for freeform optical design based on Hamilton’s characteristic functions, which can be either expressed in terms of a cost function or a generating function. The models are closed with an energy balance. We give an example of an inverse design for a near-field optical system.

Carbon dioxide is an important parameter for indoor air quality (IAQ) monitoring and demand controlled ventilation (DCV). Usually, CO2 sensors are wall‐mounted at 0.9–1.8 m (3–6 ft) height as prescribed by LEED, although ASHRAE standards seemed to relax this requirement. In this work, we investigate whether positioning these sensors in the ceiling is effective and advantageous. We studied CO2‐level measurements for HVAC control in configurations with mixing ventilation and found that CO2 from human exhalations experiences buoyancy from several factors. We calculated buoyancy from air properties, and we introduced the notion of “stratification temperature” for exhaled air. By simulation, we test the sensitivity to temperature, and we conducted in situ in vivo measurements to acquire more detailed insights in the feasibility of ceiling sensor positions. Buoyancy calculations show that in exhaled air, the positive buoyancy of H2O approximately compensates for the negative buoyancy of CO2, so that thermal buoyancy is the most dominant factor. Exhaled air, containing CO2 to be measured, will rise towards a ceiling that has a temperature below the stratification temperature. Computational fluid dynamics (CFD) simulations of a small office space indicate that this can also be the case in the presence of air flows induced by a mechanical ventilation system. The measurement results support that using “properly mounted” CO2 sensors in the ceiling gives lower variability in CO2 measurements and faster response than wall‐mounted sensors and yields slightly higher values than wall sensors. Our results highlight the need to update the standards and regulations for sensing CO2 to include ceiling‐mounted sensors.

Computer vision techniques are becoming increasingly popular for monitoring pig behavior. For instance, object detection models allow us to detect the presence of pigs, their location, and their posture. The performance of object detection models can be affected by variations in lighting conditions (e.g., intensity, spectrum, and uniformity). Furthermore, lighting conditions can influence pigs' active and resting behavior. In the context of experiments testing different lighting conditions, a detection model was developed to detect the location and postures of group-housed growing-finishing pigs. The objective of this paper is to validate the model developed using YOLOv8 detecting standing, sitting, sternal lying, and lateral lying pigs. Training, validation, and test datasets included annotation of pigs from 10 to 24wk of age in 10 different light settings; varying in intensity, spectrum, and uniformity. Pig detection was comparable for the different lighting conditions, despite a slightly lower posture agreement for warm light and uneven light distribution, likely due to a less clear contrast between pigs and their background and the presence of shadows. The detection reached a mean average precision (mAP) of 89.4%. Standing was the best-detected posture with the highest precision, sensitivity, and F1 score, while the sensitivity and F1 score of sitting was the lowest. This lower performance resulted from confusion of sitting with sternal lying and standing, as a consequence of the top camera view and a low occurrence of sitting pigs in the annotated dataset. This issue is inherent to pig behavior and could be tackled using data augmentation. Some confusion was reported between types of lying due to occlusion by pen mates or pigs' own bodies, and grouping both types of lying postures resulted in an improvement in the detection (mAP = 97.0%). Therefore, comparing resting postures (both lying types) to active postures could lead to a more reliable interpretation of pigs' behavior. Some detection errors were observed, e.g., two detections for the same pig were generated due to posture uncertainty, dirt on cameras detected as a pig, and undetected pigs due to occlusion. The localization accuracy measured by the intersection over union was higher than 95.5% for 75% of the dataset, meaning that the location of predicted pigs was very close to annotated pigs. Tracking individual pigs revealed challenges with ID changes and switches between pen mates, requiring further work.