This article reviews the applications of artificial neural networks (ANNs) in greenhouse technology, and also presents how this type of model can be developed in the coming years by adapting to new technologies such as the internet of things (IoT) and machine learning (ML). Almost all the analyzed works use the feedforward architecture, while the recurrent and hybrid networks are little exploited in the various tasks of the greenhouses. Throughout the document, different network training techniques are presented, where the feasibility of using optimization models for the learning process is exposed. The advantages and disadvantages of neural networks (NNs) are observed in the different applications in greenhouses, from microclimate prediction, energy expenditure, to more specific tasks such as the control of carbon dioxide. The most important findings in this work can be used as guidelines for developers of smart protected agriculture technology, in which systems involve technologies 4.0.

The present manuscript focuses on reviewing the optical techniques proposed to monitor milk quality in dairy farms to increase productivity and reduce costs. As is well known, the quality is linked to the fat and protein concentration; in addition, this issue is crucial to maintaining a healthy herd and preventing illnesses such as mastitis and ketosis. Usually, the quality of the milk is carried out with invasive methods employing chemical reagents that increase the time analysis. As a solution, several spectroscopy optical methods have been proposed, here, the benefits such as non-invasive measurement, online implementation, rapid estimation, and cost-effective execution. The most attractive optical methods to estimate fat and protein in cow’s milk are compared and discussed considering their performance. The analysis is divided considering the wavelength operation (ultraviolet, visible, and infrared). Moreover, the weaknesses and strengths of the methods are fully analyzed. Finally, we provide the trends and a recent technique based on spectroscopy in the visible wavelength.

The present work experimentally demonstrates a multimode fiber optic sensing setup for total fat detection in raw milk samples. The optical fiber arrangement incorporates a low-coherence Fabry–Perot cavity operating in dual response. The system provides a phase modulation for a total fat range from 0.97 to 4.36%. Here, the protein remains constant at ≈3%. The data indicate that maximum sensitivity close to 616 pm/%fat could be achieved at optimal wavelength operation (500 nm). In addition, the system presented a minimal repeatability error measurement of 0.08%, cross-sensitivity between protein and fat of 0.134, and a regression coefficient of r2=0.9763. A thermal analysis was also performed, which indicate the temperature immunity of the system. The proposed method represents a low-cost alternative to detect minimal fat variations in raw cow milk.

No abstract

One of the principal activities in the poultry industry is determine the sex of chickens of one day old. In this paper, we present a non-invasive technique to determine the sex of day-old chickens based on image processing algorithms. The technique analyzes morphometrical attributes from chicken using the slow and rapid growth of primary and secondary feathers patterns and linear discriminant analysis models. Based on the area formed with the superior points of each feather, the technique is capable to determine the sex of day-old chickens with an accuracy of 94.4%, providing a cheap, non-invasive, and high accurate technique that could be implemented onto a dedicated and automated system.