Since the early beginning of the development of natural sciences, collecting and assay of huge amounts of data was one of the leading analytical tools. The same goes for environmental sciences and environmental engineering, which produce higher demand for efficient and productive approaches to work with continuously increasing sizes of the collected data from a huge variety of research fields every day. (Kendall and Costello 2006) Nowadays, machine learning algorithms have proven themselves as a universal tool for different types of tasks, giving advanced possibilities for dealing with analysed data, including such types of tasks as data imputation, unsupervised clusterization, classification and regression. They are commonly used in many research areas; however, they are yet less common among environmental engineering workers, though such tools may provide an extremely efficient alternative to the traditional analytical approaches. (Wilcox, Woon and Aung 2013) The purpose of the research behind this thesis was in presenting of examples of how such advanced tools may be used on a particular data set meant for increasing water quality in European region. In the following chapters one will go through the presentation of the machine learning, it’s origins and possibilities in general, explanation of the data and models used during the research, results of the application of algorithms, discussion (covering obstacles one can face while working with this kind of models) and conclusion, which will cover the presented material, give advices for engineers and scientists who would like to use this models for their environmental tasks and finally and give some words about the possible future of the development of these tools in environmental field.
Its main purpose is to facilitate information exchange regarding the waste to be collected from individuals or from waste collection points, thereby exploiting the wide acceptance and use of smartphones. To improve waste collection planning, individuals would photograph the waste item and upload the image to the waste colle2ction company server, where it would be recognized and classified automatically. The proposed system can be operated on a server or through a mobile app. A novel method of classification and identification using neural networks is proposed for image analysis: a deep learning convolutional neural network (CNN) was applied to classify the type of e-waste, and a faster region-based convolutional neural network (R-CNN) was used to detect the category and size of the waste equipment in the images.
Our aim of the project is to make use of train and test sets to predict the material given like it is plastic or metal or glass etc.
A Convolutional Neural Network (ConvNet/CNN) is a Deep Learning algorithm which can take in an input image, assign importance (learnable weights and biases) to various aspects/objects in the image and be able to differentiate one from the other. A test set and train set are used to validate the model.
The accumulation of solid waste in the urban area is becoming a great concern, and it would result in environmental pollution and may be hazardous to human health if it is not properly managed. It is important to have an advanced/intelligent waste management system to manage a variety of waste materials.
The present way of separating waste/garbage is the hand-picking method, whereby someone is employed to separate out the different objects/materials. The person, who separate waste, is prone to diseases due to the harmful substances in the garbage. With this in mind, it motivated us to develop an automated system which is able to sort the waste and this system can take short time to sort the waste, and it will be more accurate in sorting than the manual way. With the system in place, the beneficial separated waste can still be recycled and converted to energy and fuel for the growth of the economy. The system that is developed for the separation of the accumulated waste is based on the combination of Convolutional Neural Network with recognition and classification.
It shows whether the chosen image is Glass, Metal, Cardboard, Trash, Plastic, Paper.
Advantages: The main advantage of CNN compared to its predecessors is that it automatically detects the important features without any human supervision. For example, given many pictures of cats and dogs, it can learn the key features for each class by itself
Disadvantages: CNN do not encode the position and orientation of object. Lack of ability to be spatially invariant to the input data
The classification of trash within the scope of recycling is possible with machine learning methods. Further data are needed to achieve higher accuracy rates. In the context of our proposed model, we have achieved high classification success without using any method of data augmentation. Studies show that the number of images and classes in the data set can be increased and a more comprehensive recycling project can be realized
It can reduce the man power and can also can prevent humans and animals in spreading of diseases due to the waste material