Developing plastic recycling classifier by deep learning and directed acyclic graph residual network

Abstract

Recycling is one of the most important approaches to safeguard the environment since it aims to reduce waste in landfills while conserving natural resources. Using deep Learning networks, this group of wastes may be automatically classified on the belts of a waste sorting plant. However, a basic set of connected layers may not be adequate to give satisfactory accuracy for such multi output classifier tasks. To optimize the gradient flow and enable deeper training for network design with multi label classifier, this study suggests a residual-based deep learning convolutional neural network. For network training, ten classes have been explored. The Directed Acyclic Graph (DAG) is a structure with hidden layers that have inputs, outputs, and other layers. The DAG network's residual-based architecture features shortcut connections that bypass some levels of the network, allowing gradients of network parameters to travel freely among the network output layers for deeper training. The methodology includes: 1) preparing the data and creating an augmented image data store; 2) defining the main serially-connected branches of the network architecture; 3) defining the residual interconnections that bypass the main branch layers; 4) defining layers, and finally; 5) creating a residual-based deeper layer graph. The concept is to split down the multiclass classification problem into minor binary states, where every classifier performs as an expert by concentrating on discriminating between only two labels, improving total accuracy. The results achieve (2.861 %) training error and (9.76 %) a validation error. The training results of this classifier are evaluated by finding the training error, validation error, and showing the confusion matrix of validation data

Developing three dimensional localization system using deep learning and pre-trained architectures for IEEE 802.11 Wi-Fi

“…The labels and features combination is used to train the Convolutional Neural Network (CNN) for the location names and STA positions [9]. Dataset is created initially for the deep learning CNN stage to estimate functions over varied domain ranges.…”

“…The user location estimates of the network are based on received signals of unknown locations and rely on a reference database [6][7][8]. Directed Acyclic Graph residual network of Deep Learning was widely used for image classification [9] and for improving noisy images that are already filtered by the bilateral process via a multi-scale context aggregation network as discussed in [10].…”

Evaluating image encryption algorithms for the hyperchaotic system and fibonacci q-matrix, secure internet of things, and advanced encryption standard

“…Due to the expanding use of images in a variety of fields, it is crucial to safeguard sensitive picture data from unwanted access. Such encryption algorithms also can contribute to adding secure monitoring of industrial processing such that in [15]. There are always unresolved issues regarding the analysis on image encryption processes especially for recent encryption techniques such as the Fibonacci Q-matrix in hyperchaotic to be evaluated and compared according to merit criteria on image encryption.…”

Developing satellite hyperspectral image processing using a maximum abundance classifier with nine ground truth classes

“…

of bands of the photographed substance, hyperspectral imaging delivers a great density of spectral data [4,5]. The majority of contemporary hyperspectral sensors also have a high spatial resolution, allowing the images to be used for a variety of purposes, including agriculture, geosciences, biomedical imaging, molecular biology, astronomy, and surveillance.Therefore, research and further analysis on the development of processing the hyperspectral images to determine the proportion of material or end-member contributions in each pixel, making it beneficial for the identification or detection of materials are relevant.

…”

“…of bands of the photographed substance, hyperspectral imaging delivers a great density of spectral data [4,5]. The majority of contemporary hyperspectral sensors also have a high spatial resolution, allowing the images to be used for a variety of purposes, including agriculture, geosciences, biomedical imaging, molecular biology, astronomy, and surveillance.…”