Waste management is also an important concern in the modern world, which is fast urbanizing and industrializing. The conventional waste sorting techniques are ineffective, expensive, and dangerous to health. The current study will address these issues by suggesting a new Waste Sorting System with Deep Learning (enabled with the use of Wireless IoTs). The first aim is to improve the quality and performance of waste classification. The model suggested is a combination of Convolutional neural networks (CNNs) and IoTs to enable real-time data processing and sorting automatically. CNN model is a model that is purposefully suited to large, high-dimensional data sets, which capture intricate patterns in waste photographs. The evaluation of the system was conducted based on a comprehensive dataset of six waste types such as cardboard, glass, metal, paper, plastic, and trash. The outcomes show that there is a high accuracy of 56% and specifically high on cardboard classification 0.86 and high recall of paper 0.88. This data has shown that the suggested model is much more effective in terms of the accuracy of classification and efficiency of operating than the conventional ones. Although the model has certain weaknesses in differentiating between similar categories (that should be visually similar), it offers a strong basis on which improvements can be made in the future. The study contributes to the body of research on waste management sustainability because it provides a scalable, efficient, and automated system to sort waste, and thus introduce additional innovation in the field.
Latha et al. (Wed,) studied this question.