Public authorities and private companies have used video cameras as part of surveillance systems, and one of their objectives is the rapid detection of physically violent actions. This task is usually performed by human visual inspection, which is labor-intensive. For this reason, different deep learning models have been implemented to remove the human eye from this task, yielding positive results. One of the main problems in detecting physical violence in videos is the variety of scenarios that can exist, which leads to different models being trained on datasets, leading them to detect physical violence in only one or a few types of videos. In this work, we present an approach for physical violence detection on images obtained from video based on threshold active learning, that increases the classifier’s robustness in environments where it was not trained. The proposed approach consists of two stages: In the first stage, pre-trained neural network models are trained on initial datasets, and we use a threshold () to identify those images that the classifier considers ambiguous or hard to classify. Then, they are included in the training dataset, and the model is retrained to improve its classification performance. In the second stage, we test the model with video images from other environments, and we again employ () to detect ambiguous images that a human expert analyzes to determine the real class or delete the ambiguity on them. After that, the ambiguous images are added to the original training set and the classifier is retrained; this process is repeated while ambiguous images exist. The model is a hybrid neural network that uses transfer learning and a threshold to detect physical violence on images obtained from video files successfully. In this active learning process, the classifier can detect physical violence in different environments, where the main contribution is the method used to obtain a threshold (which is based on the neural network output) that allows human experts to contribute to the classification process to obtain more robust neural networks and high-quality datasets. The experimental results show the proposed approach’s effectiveness in detecting physical violence, where it is trained using an initial dataset, and new images are added to improve its robustness in diverse environments.
The increasing pollution of surface waters necessitates continuous monitoring of key environmental parameters to assess water quality. Dissolved Oxygen (DO) is a critical indicator of the health of aquatic ecosystems and is influenced by factors such as temperature, pH, and electrical conductivity. Traditional water quality models often require extensive datasets, which can imply challenges for large-scale monitoring. In this study, we evaluate the effectiveness of Artificial Neural Networks (ANNs) in predicting DO levels by comparing seven different ANN architectures to nine classical regression models. We utilize data from the Water Quality Prediction dataset available in the University of California at Irvine (UCI) Machine Learning Repository, which covers 37 geographic regions across the United States. Our analysis shows that a Multi-Layer Perceptron (MLP) outperforms traditional regression approaches. The MLP effectively captures complex, nonlinear relationships in water quality data, achieving higher predictive accuracy as measured by the coefficient of determination (R²). These findings emphasize the potential of deep learning methodologies as reliable alternatives to conventional statistical models for water quality prediction, offering improved accuracy and efficiency in environmental monitoring.
Automated violence detection in images presents a technical and scientific challenge that demands specialized methods to enhance classification systems. This study introduces an approach for automatically identifying relevant samples to improve the performance of neural network models, specifically DenseNet121, with a focus on violence classification in images. The proposed methodology begins with an initial training phase using a balanced dataset (DS1, 6000 images). Based on the model’s output scores (), three confidence levels are defined: Safe ( or ), Border (), and Average (). These levels correspond to scenarios with low, moderate, and high prediction error probabilities, respectively, where is an adjustable threshold. The Border subset exhibits symmetry around the decision boundary (), capturing maximally uncertain samples, while the Safe regions reflect functional asymmetries in high-confidence predictions. Subsequently, these thresholds are applied to a second dataset (DS2, 5600 images) to extract specialized subsets for retraining (, , and ). Finally, the model is evaluated using an independent test set (, 4400 images), ensuring complete data isolation. The experimental results demonstrate that the confidence-based subsets offer competitive performance despite using significantly fewer samples. The Average subset achieved an F1-Score of 0.89 and a g-mean of 0.93 using only 20% of the data, making it a promising alternative for efficient training. These findings highlight that strategic sample selection based on confidence thresholds enables effective training with reduced data, offering a practical balance between performance and efficiency when symmetric uncertainty modeling is exploited.