Edgar Scavino

Intelligent computer vision system for segregating recyclable waste papers

This article explores the application of image processing techniques in recyclable waste paper sorting. In recycling, waste papers are segregated into various grades as they are subjected to different recycling processes. Highly sorted paper streams facilitate high quality end products and save processing chemicals and energy. From 1932 to 2009, different mechanical and optical paper sorting methods have been developed to fill the paper sorting demand. Still, in many countries including Malaysia, waste papers are sorted into different grades using a manual sorting system. Because of inadequate throughput and some major drawbacks of mechanical paper sorting systems, the popularity of optical paper sorting systems has increased. Automated paper sorting systems offer significant advantages over human inspection in terms of worker fatigue, throughput, speed, and accuracy. This research attempts to develop a smart vision sensing system that is able to separate the different grades of paper using first-order features. To construct a template database, a statistical approach with intra-class and inter-class variation techniques are applied to the feature selection process. Finally, the K-nearest neighbor (KNN) algorithm is applied for paper object grade identification. The remarkable achievement obtained with the method is the accurate identification and dynamic sorting of all grades of papers using simple image processing techniques.

Recyclable Waste Paper Sorting Using Template Matching

This paper explores the application of image processing techniques in recyclable waste paper sorting. In recycling, waste papers are segregated into various grades as they are subjected to different recycling processes. Highly sorted paper streams will facilitate high quality end products, and save processing chemicals and energy. Since 1932 to 2009, different mechanical and optical paper sorting methods have been developed to fill the demand of paper sorting. Still, in many countries including Malaysia, waste papers are sorted into different grades using manual sorting system. Due to inadequate throughput and some major drawbacks of mechanical paper sorting systems, the popularity of optical paper sorting systems is increased. Automated paper sorting systems offer significant advantages over human inspection in terms of fatigue, throughput, speed, and accuracy. This research attempts to develop a smart vision sensing system that able to separate the different grades of paper using Template Matching. For constructing template database, the RGB components of the pixel values are used to construct RGBString for template images. Finally, paper object grade is identified based on the maximum occurrence of a specific template image in the search image. The outcomes from the experiment in classification for White Paper, Old Newsprint Paper and Old Corrugated Cardboard are 96%, 92% and 96%, respectively. The remarkable achievement obtained with the method is the accurate identification and dynamic sorting of all grades of papers using simple image processing techniques.

Object identification using DNA computing algorithm

Although template matching method is widely used for object identification, the large computational time is the major drawback to use this method in real time application. Using the concepts of replication and massive parallelism operations, the DNA computing algorithm can efficiently reduce the computational time of the template matching method. The emphasis of this research has been given in two objectives, namely development of a generic DNA computing algorithm for object identification based on the theme of the template matching technique and application of this algorithm for recyclable waste paper sorting. The achieved classification success rates are 92%, 90%, and 93% with template size 5 × 5 pixels for white paper, old newsprint paper and old corrugated cardboard, respectively.

Real time road sign recognition system using artificial neural networks for bengali textual information box

An Automated Road Sign Recognition system using Artificial Neural Network for the Textual Information box inscribing in Bengali is presented in this paper. The system captures real time images every two seconds and saves them as JPG format files. The system processes the images to find out whether they contain images of road signs or not. The textual information of the road signs is detected and extracted from the images. The Bengali OCR system takes the textual information as an input to recognize individual Bengali characters. The Bengali OCR is implemented using Multi layer Perceptron. The output of the Bengali OCR system is compared with the previously enrolled standard Bengali textual road signs. The throughput which comes from the matching process is used as input for the speech synthesizer and finally the system delivers the audio stream to the driver, either in Bengali or in English based on the user settings. After testing this system, the obtained accuracy rate was evaluated at 91.48%.

Backtracking search algorithm in CVRP models for efficient solid waste collection and route optimization

Waste collection is an important part of waste management that involves different issues, including environmental, economic, and social, among others. Waste collection optimization can reduce the waste collection budget and environmental emissions by reducing the collection route distance. This paper presents a modified Backtracking Search Algorithm (BSA) in capacitated vehicle routing problem (CVRP) models with the smart bin concept to find the best optimized waste collection route solutions. The objective function minimizes the sum of the waste collection route distances. The study introduces the concept of the threshold waste level (TWL) of waste bins to reduce the number of bins to be emptied by finding an optimal range, thus minimizing the distance. A scheduling model is also introduced to compare the feasibility of the proposed model with that of the conventional collection system in terms of travel distance, collected waste, fuel consumption, fuel cost, efficiency and CO2 emission. The optimal TWL was found to be between 70% and 75% of the fill level of waste collection nodes and had the maximum tightness value for different problem cases. The obtained results for four days show a 36.80% distance reduction for 91.40% of the total waste collection, which eventually increases the average waste collection efficiency by 36.78% and reduces the fuel consumption, fuel cost and CO2 emission by 50%, 47.77% and 44.68%, respectively. Thus, the proposed optimization model can be considered a viable tool for optimizing waste collection routes to reduce economic costs and environmental impacts.

Capacitated vehicle-routing problem model for scheduled solid waste collection and route optimization using PSO algorithm

Waste collection widely depends on the route optimization problem that involves a large amount of expenditure in terms of capital, labor, and variable operational costs. Thus, the more waste collection route is optimized, the more reduction in different costs and environmental effect will be. This study proposes a modified particle swarm optimization (PSO) algorithm in a capacitated vehicle-routing problem (CVRP) model to determine the best waste collection and route optimization solutions. In this study, threshold waste level (TWL) and scheduling concepts are applied in the PSO-based CVRP model under different datasets. The obtained results from different datasets show that the proposed algorithmic CVRP model provides the best waste collection and route optimization in terms of travel distance, total waste, waste collection efficiency, and tightness at 70-75% of TWL. The obtained results for 1 week scheduling show that 70% of TWL performs better than all node consideration in terms of collected waste, distance, tightness, efficiency, fuel consumption, and cost. The proposed optimized model can serve as a valuable tool for waste collection and route optimization toward reducing socioeconomic and environmental impacts.

DNA computer based algorithm for recyclable waste paper segregation

DNA computer based algorithm is developed and evaluated for recyclable waste paper segregation.The concepts of replication and massive parallelism operations are used.The matching stage consists of Copy, Extract, Detect and Merge or Union operations.Gel electrophoresis operation is used to identify the candidate paper object grade.Success rate are 92% for WP, 90% for ONP and 93% for OCC with template size 5i?5. This article explores the application of DNA computing in recyclable waste paper sorting. The primary challenge in paper recycling is to obtain raw materials with the highest purity. In recycling, waste papers are segregated according to their various grades, and these are subjected to different recycling processes. Highly sorted paper streams facilitate high quality end products, while saving on processing chemicals and energy. In the industry, different sensors are used in paper sorting systems, namely, ultrasonic, lignin, gloss, stiffness, infra-red, mid-infra red, and color sensors. Different mechanical and optical paper sorting systems have been developed based on the different sensors. However, due to inadequate throughput and some major drawbacks related to mechanical paper sorting systems, the popularity of optical paper sorting systems has increased. The automated paper sorting systems offer significant advantages over the manual systems in terms of human fatigue, throughput, speed, and accuracy. This research has two objectives: (1) to use a web camera as an image sensor for the vision system in lieu of different sensors; and (2) to develop a new DNA computing algorithm based on the theme of template matching techniques for segregating recyclable waste papers according to paper grades. Using the concepts of replication and massive parallelism operations, the DNA computing algorithm can efficiently reduce the computational time of the template matching method. This is the main strength of the DNA computing algorithm in actual inspections. The algorithm is implemented by using a silicon-based computer to verify the success rate in paper grade identification.

An efficient segmentation technique for known touching objects using a genetic algorithm approach

This paper presents a genetic algorithm (GA) based segmentation technique that can separate two touching objects intended for an automatic recognition of plastic bottles moving on a conveyor belt. The proposed method is based on the possibility to separate the two objects by means of a straight line, whose position is determined by a GA. Extensive testing shows that the proposed method is fast and yields high success rate of correct segmentation with only a limited number of both chromosomes and iterations.

The design of a complete uniform segmented display unit for Arabic alphanumeric characters

There are four forms for most of the Arabic characters- initial, medial, final and isolated. This paper presents the design of a complete display unit for Arabic alphanumeric characters using uniform segments. The word ‘complete’ signifies that the proposed display unit can represent all forms of entire Arabic alphabets as well as Arabic numerals. The word ‘uniform’ is used because of avoiding curved segments. The first segmented display unit for Arabic characters in the world covered only the isolated form and did not follow any base line to display a sequence of characters. This work extends the capability to represent all the forms of Arabic characters and follows a base line to represent a sequence of characters. Here a grid structure of 25 non-overlapping segments has been discovered. There can be 109 alphanumeric characters (at least 99 characters such as isolated form- 30, final form-29, medial form- 21, initial form- 20; and 10 numerals) to be displayed and 7-bit input is used to represent each character. After analyzing appropriate segments to be activated for every character, respective logic functions of all segments have been derived to display the entire set of Arabic alphanumeric characters.