The rapid growth of the world’s population has put significant pressure on agriculture to meet the increasing demand for food. In this context, agriculture faces multiple challenges, one of which is weed management. While herbicides have traditionally been used to control weed growth, their excessive and random use can lead to environmental pollution and herbicide resistance. To address these challenges, in the agricultural industry, deep learning models have become a possible tool for decision-making by using massive amounts of information collected from smart farm sensors. However, agriculture’s varied environments pose a challenge to testing and adopting new technology effectively. This study reviews recent advances in deep learning models and methods for detecting and classifying weeds to improve the sustainability of agricultural crops. The study compares performance metrics such as recall, accuracy, F1-Score, and precision, and highlights the adoption of novel techniques, such as attention mechanisms, single-stage detection models, and new lightweight models, which can enhance the model’s performance. The use of deep learning methods in weed detection and classification has shown great potential in improving crop yields and reducing adverse environmental impacts of agriculture. The reduction in herbicide use can prevent pollution of water, food, land, and the ecosystem and avoid the resistance of weeds to chemicals. This can help mitigate and adapt to climate change by minimizing agriculture’s environmental impact and improving the sustainability of the agricultural sector. In addition to discussing recent advances, this study also highlights the challenges faced in adopting new technology in agriculture and proposes novel techniques to enhance the performance of deep learning models. The study provides valuable insights into the latest advances and challenges in process systems engineering and technology for agricultural activities.

Food waste occurs from harvesting to consumption. Applying procedures and technologies, changing attitudes, and promoting awareness have positive social, economic, and environmental impacts that can contribute to reducing food waste. The paper presents a decision support system (DSS) to predict the quality evolution of fruits and vegetables, particularly of peaches, and estimate its commercialization period at the highest overall perceived quality by consumers, thus contributing to reducing food waste. The Fuzzy Logic DSS predicts the evolution of the physical-chemical parameters of peaches (hardness, soluble solids content, and acidity) depending on the cultivar (Royal Summer and Royal Time), storage time, and temperature. As the range of the values of these physical-chemical parameters of peaches that consumers perceive to be at their highest quality are known, the DSS predicts the marketable period in days. Case studies were developed to analyze the influence of each physical-chemical parameter on the commercialization days (number and time to start). It is concluded that temperature is the most important parameter for fruit conservation. A low value of conservation temperature allows for the significant extension of the time that peaches can be sold at the highest quality. Hardness is used to determine the harvest date since it is an index of fruit ripeness. The same conclusion is obtained for the influence of the soluble solids content. The influence of acidity on marketable days is less than the other physical-chemical parameters. This DSS helps retailers to sell their peaches at the highest quality with benefits for all parties. It also helps in the decision-making concerning the actions to take when fruits are reaching the end of their highest quality by predicting the range of the commercialization days. This formulation can be extended to other fruits and vegetables and in the last instance contribute to the reduction of food loss and waste, consequently promoting social, economic, and environmental aspects of our daily life.

The global population’s rapid growth necessitates a 70% increase in agricultural production, posing challenges exacerbated by weed infestation and herbicide drawbacks. To address this, machine learning (ML) models, particularly convolutional neural networks (CNNs), are employed in precision agriculture (PA) for weed detection. This study focuses on testing CNN architectures for image classification tasks using the PyTorch framework, emphasizing hyperparameter optimization. Four groups of experiments were carried out: the first one trained all the PyTorch architectures, followed by the creation of a baseline, the evaluation of a new and extended dataset in the best models, and finally, the test phase was conducted using a web application developed for this purpose. Of 80 CNN sub-architectures tested, the MaxVit, ShuffleNet, and EfficientNet models stand out, achieving a maximum accuracy of 96.0%, 99.3%, and 99.3%, respectively, for the first test phase of PyTorch classification architectures. In addition, EfficientNet_B1 and EfficientNet_B5 stood out compared to all other models. During experiment 3, with a new dataset, both models achieved a high accuracy of 95.13% and 94.83%, respectively. Furthermore, in experiment 4, both EfficientNet_B1 and EfficientNet_B5 achieved a maximum accuracy of 96.15%, the highest one. ML models can help to automate crop problem detection, promote organic farming, optimize resource use, aid precision farming, reduce waste, boost efficiency, and contribute to a greener, sustainable agricultural future.

The adoption of wireless medical technologies in hospital environments is often limited by cellular coverage issues, especially in indoor areas with complex structures. This study presents a detailed radio spectrum measurement campaign conducted at the Cova da Beira University Hospital Center (CHUCB), using the NARDA SRM-3006 and R&S^®TSME6 equipment. The signal strength and quality of 5G NR, LTE, UMTS, and NB-IoT technologies were evaluated. Critical coverage gaps were identified, particularly at points 17, 19, and 21. Results revealed that operators MEO and NOS dominate coverage, with MEO providing better 5G NR coverage and NOS excelling in LTE signal quality. Based on the results, the localized installation of femtocells is proposed to improve coverage in these areas. The approach was designed to be scalable and replicable, with a planned application at Cumura Hospital (Guinea-Bissau), reinforcing the applicability of the solution in contexts with limited infrastructure. This work provides both technical and clinical contributions to achieving ubiquitous cellular coverage in healthcare settings.