Anomaly detection is a challenging problem in various application domains of Artificial Intelligence, such as in video surveillance, the Internet of Things, and notably, precision agriculture. The effectiveness of anomaly detection in each field is intricately linked to the domain-specific data, adhering, at the same time, to the core objective of detecting outliers. In the precision agriculture domain, anomalies range from plant diseases in image data to fluctuating environmental conditions in time-series datasets. This review provides a detailed examination of deep learning-based anomaly detection methods within precision agriculture, adopting the PRISMA methodology for a structured and comprehensive analysis. We employ a novel taxonomy categorizing recent literature by agricultural application, anomaly relevance, data modality, deep learning architecture, supervision level, and dataset usage. Our findings highlight a predominant reliance on visual data and uncover a potential alignment between methods originally devised for classification or detection and the anomaly detection challenge. The review also signals a pressing need for large-scale datasets to address precision agriculture challenges effectively. By mapping the current landscape and suggesting directions for future research, our work aims to facilitate advancements in anomaly detection techniques, enabling enhanced decision-making and operational efficiency in precision agriculture.

This work examines violence detection in video scenes of crowds and proposes a crowd violence detection framework based on a 3D convolutional deep learning architecture, the 3D-ResNet model with 50 layers. The proposed framework is evaluated on the Violent Flows dataset against several state-of-the-art approaches and achieves higher accuracy values in almost all cases, while also performing the violence detection activities in (near) real-time.

Surveillance systems currently deploy a variety of devices that can capture visual content (such as CCTV, body-worn cameras, and smartphone cameras), thus rendering the monitoring of video footage obtained from multiple such devices a complex task. This becomes especially challenging when monitoring social events that involve large crowds, particularly when there is a risk of crowd violence. This paper presents and demonstrates a crowd violence detection system that can process, analyze, and alert potential stakeholders, when violence-related content is identified in crowd-based video footage. Based on deep neural networks, the proposed end-to-end framework utilizes a 3D Convolutional Neural Network (CNN) to deal with the (near) real-time analysis of video streams and video files for crowd violence detection. The framework is trained, evaluated, and demonstrated using the Violent Flows dataset, a dataset related to crowd violence that is widely used for research. The presented framework is provided as a standalone application for desktop environments and can analyze both video streams and video files.

This work presents a spatio-temporal activity detection and recognition framework for untrimmed surveillance videos consisting of a three-step pipeline: object detection, tracking, and activity recognition. The framework relies on the YOLO v4 architecture for object detection, Euclidean distance for tracking, while the activity recognizer uses a 3D Convolutional Deep learning architecture employing spatio-temporal boundaries and addressing it as multi-label classification. The evaluation experiments on the VIRAT dataset achieve accurate detections of the temporal boundaries and recognitions of activities in untrimmed videos, with better performance for the multi-label compared to the multi-class activity recognition.