The pursuit of sustainable energy production through the conversion of agricultural waste into different bioenergy resources is of paramount importance given its potential to mitigate environmental impact while meeting energy demands. In this review, a comprehensive overview of the technologies for the biochemical and thermochemical conversion of agricultural waste into bioenergy is provided. A summary of the process of its conversion into different bioenergy products such as biogas, bio-oil, and biofuel is provided, in addition to the potential advantages and challenges faced using different biomass conversion technologies. The review highlights the potential of agricultural waste valorization to address the current energy demand while at the same time contributing to environmental benefits and greenhouse gas emission reductions. Moreover, this review highlights some significant gaps for improvement. These include the challenges in the pretreatment of agricultural waste biomass in optimizing the conversion rates and lowering the required energy consumption throughout the process while enhancing both the quantity and quality of the output. Some recommendations are proposed to address the identified challenges. These include the need for further studies for a thorough assessment to evaluate the efficacity and sustainability of agricultural waste valorization technologies. Assessment methods such as life cycle assessment (LCA), life cycle analysis (LCA), net energy ratio (NER) calculations, life cycle costing (LCC), as well as techno-economic assessment (TEA), are recommended, together with collaboration among governments, farmers, and researchers, as well as the integration of cutting-edge technologies to enhance various aspects of agricultural waste, optimizing the conversion process, cost efficiency, time management, and labor requirements, consequently boosting the conversion efficiency and product quality.

Automated detection of dead chickens is essential for enhancing biosecurity, animal welfare, and operational efficiency in poultry farms. This study evaluates the performance of YOLOv8n, YOLOv9c, YOLOv10n, and YOLOv11n for detecting dead chickens in cage-free poultry farms. A synthetic dataset of 3413 images was created by compositing manually annotated images of dead and healthy chickens into realistic stall backgrounds to simulate real farm conditions. The models were assessed using standard object detection metrics (precision, recall, and mean average precision (mAP) at IoU thresholds of 0.5 and 0.5-0.95) alongside computational efficiency indicators including inference speed, frames per second (FPS), model size, and training time. YOLOv9c achieved the highest detection accuracy (mAP@50 = 0.983, mAP@50-95 = 0.93), making it the most reliable for minimising false positives and missed detections. YOLOv11n delivered the fastest inference speed (2.8 ms/frame, ∼357 FPS), making it more suitable for real-time applications. These results underscore the importance of selecting a YOLO model based on farm-specific operational constraints. YOLOv9c is recommended for accuracy-critical tasks, YOLOv11n for real-time monitoring, and YOLOv8n or YOLOv10n for resource-limited deployments. Comparative analysis with earlier YOLO models (YOLOv3-YOLOv7) shows that newer versions improve both detection reliability and processing speed. This work contributes a performance benchmark to guide AI-based poultry monitoring and highlights future directions, including real-world deployment and validation under live farm conditions.

The goal of this research was to implement machine vision algorithms in a cattle stable to detect cattle in stalls and determine their activities. It also focused on finding the optimal hyperparameter settings for training the model, as balancing prediction accuracy, training time, and computational demands is crucial for real-world implementation. The investigation of suitable parameters was carried out on the YOLOv5 convolutional neural network (CNN). The types of the YOLOv5 network (v5x, v5l, v5m, v5s, and v5n), the effect of the learning rate (0.1, 0.01, and 0.001), the batch size (4, 8, 16, and 32), and the effect of the optimizer used (SGD and Adam) were compared in a step-by-step process. The main focus was on mAP 0.5 and mAP 0.5:0.95 metrics and total training time, and we came to the following conclusions: In terms of optimization between time and accuracy, the YOLOv5m performed the best, with a mAP 0.5:0.95 of 0.8969 (compared to 0.9070 for YOLOv5x). The training time for YOLOv5m was 7:48:19, while YOLOv5x took 16:53:27. When comparing learning rates, the variations in accuracy and training time were minimal. The highest accuracy (0.9028) occurred with a learning rate of 0.001, and the lowest (0.8897) with a learning rate of 0.1. For training time, the fastest was 7:47:17, with a difference of 1:02:00 between the fastest and slowest times. When comparing the effect of batch size, model accuracy showed only minimal differences (in tenths of a percentage), but there were significant time savings. When using a batch size of 8, the training time was 12:50:48, while increasing the batch size to 32 reduced the training time to 6:07:13, thus speeding up the training process by 6:43:35. The last parameter compared was the optimizer. SGD and Adam optimizers were compared. The choice of optimizer had a minimal impact on the training time, with differences only in seconds. However, the accuracy of the trained model was 6 per cent higher (0.8969) when using the SGD optimizer.

Ammonia (NH₃) emissions have a negative impact on the welfare of breeding animals, human health, and the environment. These influences of modern intensive agriculture have led to numerous protocols, national regulations, and European Directives. Following previous regulatory measures, the Commission Implementing Decision European Union (EU) 2017/302 on 15 February 2017 has established best available technique (BAT) conclusions, under Directive 2010/75/EU of the European Parliament and the Council, for the intensive rearing of poultry and pigs. This applies to intensive poultry and pig producers with a capacity of over 40,000 poultry, 750 sows, or 2000 fattening pigs. Due to the application of this directive, air emissions have been reduced by between 40% and 75% over the last 15 years. The integrated permit monitors the entire environmental burden of the farm on its surroundings (air pollution, water, soil pollution, waste production, energy use). This review aims to provide a critical overview of how member states (including the United Kingdom) are approaching the implementation of IPPC (Integrated Pollution Prevention and Control) and the conclusions of BAT in their legislation and related documents, and how they monitor NH₃ emissions from intensive livestock farming. The data for this review were obtained from 2019 to 2020.

Cattle identification systems are advancing to meet the growing demands of precision livestock management, traceability, and ethical animal treatment. This study investigates three methods: body texture recognition, QR code collars, and numerical labelling, implemented using the YOLOv8 convolutional neural network. Each method was evaluated in terms of accuracy, scalability, adaptability to dynamic herd changes, and operational efficiency under various environmental conditions. Body texture recognition, while leveraging unique natural patterns and achieving a mean Average Precision (mAP50-95) of 0.78 proved limited by its reliance on frequent dataset retraining to accommodate changes in herd composition and susceptibility to misidentification in larger herds. QR code collars demonstrated adaptability in dynamic herds by enabling pre-trained convolutional neural networks to assign reserved codes to new animals without retraining, while removing animals involves simply deleting their codes from the system. This approach also achieved an mAP50-95 of 0.71, which was lower than the body texture-based approach, but offered greater flexibility in herd management. Despite this adaptability, this method demonstrated significant challenges in real-world environments. Occlusion caused by feeders, barriers, or animal movements, along with low-resolution imaging and poor lighting conditions, can compromise detection accuracy, particularly in larger herds with obstructive barn layouts. The numerical labelling method emerged as the most effective solution to dynamic cattle identification, achieving the highest mAP50-95 of 0.84. It provided a scalable and highly accurate approach that integrates seamlessly with automated systems. Unlike traditional body marking techniques such as ear notching and branding, numerical labelling is less invasive, painless, and highly scalable, aligning with ethical livestock management practices while maintaining consistent accuracy across diverse environmental conditions.