Deep Learning Based Smoking Detection System For Public Areas

Cigarette smoking continues to be a major worldwide health concern, with multiple negative implications for both people and society. Effective surveillance and early identification of smokers are critical steps in tackling this public health issue. This study employs deep learning CNN architecture to create a robust and accurate system for recognising cigarette smokers in public areas. In the research, a large collection of face photos was obtained from both smokers and nonsmokers, assuring variety in age, gender, and ethnicity. For feature extraction and classification, we used the convolutional neural network (CNN) architecture, a well-established deep learning model. The experimental findings illustrate the efficiency of the suggested technique, with a remarkable accuracy rate of 90% in discriminating cigarette smokers from non-smokers. Furthermore, the model's performance was assessed using important measures such as accuracy, recall, and F1-score, demonstrating its ability to balance false positives and false negatives.