Search

• Automated multi-stage method employed to classify urban wastes. • A novel model named GCDN-Net is proposed as the garbage classifier. • Four different types of waste have been classified using the garbage classifier. • Saliency map-based visualization with ScoreCAM is shown as a qualitative evaluation. The escalating waste volume due to urbanization and population growth has underscored the need for advanced waste sorting and recycling methods to ensure sustainable waste management. Deep learning models, adept at image recognition tasks, offer potential solutions for waste sorting applications. These models, trained on extensive waste image datasets, possess the ability to discern unique features of diverse waste types. Automating waste sorting hinges on robust deep learning models capable of accurately categorizing a wide range of waste types. In this study, a multi-stage machine learning approach is proposed to classify different waste categories using the "Garbage In, Garbage Out" (GIGO) dataset of 25,000 images. The novel Garbage Classifier Deep Neural Network (GCDN-Net) is introduced as a comprehensive solution, adept in both single-label and multi-label classification tasks. Single-label classification distinguishes between garbage and non-garbage images, while multi-label classification identifies distinct garbage categories within single or multiple images. The performance of GCDN-Net is rigorously evaluated and compared against state-of-the-art waste classification methods. Results demonstrate GCDN-Net's excellence, achieving 95.77% accuracy, 95.78% precision, 95.77% recall, 95.77% F1-score, and 95.54% specificity when classifying waste images, outperforming existing models in single-label classification. In multi-label classification, GCDN-Net attains an overall Mean Average Precision (mAP) of 0.69 and an F1-score of 75.01%. The reliability of network performance is affirmed through saliency map-based visualization generated by Score-CAM (class activation mapping). In conclusion, deep learning-based models exhibit efficacy in categorizing diverse waste types, paving the way for automated waste sorting and recycling systems that can mitigate costs and processing times. [ABSTRACT FROM AUTHOR]

• A novel Deep learning-based approach to reconstruct the EEG signal while preserving its morphology by removing the EOG artifacts. • Provide comprehensive multimodal performance analysis of EEG, ECG and PPG in detecting emotions. • Investigate a large number of time, frequency, and time–frequency features for emotion recognition. • Novel multimodal network for emotion recognition outperforming the state-of-the-art reported performance. Emotion Recognition Systems (ERS) play a pivotal role in facilitating naturalistic Human-Machine Interactions (HMI). The research has utilized a dataset with diverse physiological signals, including Electroencephalogram (EEG), Photoplethysmography (PPG), and Electrocardiogram (ECG), to detect emotions evoked by video stimuli. The study has addressed challenges with EEG data, particularly prefrontal channels contaminated by eye blink artifacts. To tackle this, a novel 1D deep learning model, MultiResUNet3p, effectively generated clean EEG signals. Extensive features have been extracted from each modality (TD, FD, TFD), and the study identified that combining 112 features from EEG and ECG achieved the highest accuracy. The emotion classification task encompassed six emotions, and the model demonstrates outstanding performance with 96.12% accuracy in binary classification (Positive vs. Negative) and 94.25% accuracy in a complex multiclass classification of six emotions (Happy, Anger, Disgust, Fear, Neutral, and Sad). This research underscores the potential of integrating multiple physiological signals and advanced techniques to significantly improve emotion recognition accuracy, particularly in real-world scenarios involving naturalistic Human-Machine Interactions (HMI). [ABSTRACT FROM AUTHOR]