Department Publications

Vehicle license plate recognition is an integral component of intelligent transportation systems. In this work, we present an embedded real-time license plate recognition system customized for developing countries. We address the challenge of handling complex, unstructured traffic scenes with diverse vehicle types while implementing the system on an embedded platform for low-cost deployment. Our method consists of license plate detection on a multi-vehicle image, followed by character recognition on the detected license plates. Both steps use lightweight convolutional neural networks to balance accuracy and efficiency. We also introduce the SL-LPR dataset of Sri Lankan road images, which contains a variety of vehicle types and traffic conditions typically seen in developing countries. On this dataset, the license plate detection and character recognition models achieved 93.6% mAP and 87.88% accuracy, respectively, and were competitive against larger models on several public datasets. To achieve real-time performance in a resource-constrained embedded environment, we applied low-bitwidth quantization using the Brevitas library and implemented FPGA acceleration for the models using the FINN framework. The end-to-end system can operate at 11.5~FPS when implemented on the Xilinx Kria KV260 platform. These results demonstrate that our system is effective for real-time license plate recognition on an embedded device, even in complex traffic scenarios. The SL-LPR dataset is available for research use at: https://github.com/sl-lpr-uom/SL-LPR.git.

With the recent advancement of Large Language Models (LLMs), environmental perception methods, and frameworks for simplifying robot functionality implementation, robotic receptionists present an exciting opportunity to minimize humanrelated errors. However, most existing humanoid robots are costly and overqualified for such tasks, as they are primarily designed for research applications. To address this, we introduce a standalone, offline, cost-effective smart mobile robot receptionist built on an edge device that efficiently performs a predefined set of tasks while utilizing minimal hardware and computational resources. Our robot receptionist can engage in smooth conversations, recognize and interact with both new and returning users, perform hand gestures, provide navigation assistance through verbal and physical guidance, and autonomously navigate while avoiding static and dynamic obstacles …

This research presents an efficient fruit monitoring and orchard management system to replace the existing paper-based manual process. The new method is based on five state-of-the-art technologies, namely Radio Frequency IDentification (RFID), Wi-Fi network, Mobile App, Coud database/server, and Web Application. This paper presents the proper integration of these technologies to provide an effective, worker-friendly, and cost-effective solution to the problem. The proposed method starts by attaching RFID tags to each tree and each fruit and registering them in the cloud database. The cloud server visualizes the status of the orchard and implements inventory management. Workers use a hand device for tasks such as bagging, spraying, and plucking. The orchard manager carries out task assignments and worker deployment on the web application. Each worker gets notified of the assigned tasks on his hand …

Traffic management is a serious problem in many cities around the world. Even the suburban areas are now experiencing regular traffic congestion. Inappropriate traffic control wastes fuel, time, and the productivity of nations. Though traffic signals are used to improve traffic flow, they often cause problems due to inappropriate or obsolete timing that does not tally with the actual traffic intensity at the intersection. Traffic intensity determination based on statistical methods only gives the average intensity expected at any given time. However, to control traffic accurately, it is required to know the real-time traffic intensity. In this research, image processing and machine learning have been used to estimate actual traffic intensity in real time. General-purpose electronic hardware has been used for in-situ image processing based on the edge-detection method. A deep neural network (DNN) was trained to infer traffic intensity in each image in real time. The trained DNN estimated traffic intensity accurately in 90% of the real-time images during road tests. The electronic system was implemented on a Raspberry Pi single-board computer; hence, it is cost-effective for large-scale deployment.

Recent work done on traffic sign and traffic light detection focus on improving detection accuracy in complex scenarios, yet many fail to deliver real-time performance, specifically with limited computational resources. In this work, we propose a simple deep learning based end-to-end detection framework, which effectively tackles challenges inherent to traffic sign and traffic light detection such as small size, large number of classes and complex road scenarios. We optimize the detection models using TensorRT and integrate with Robot Operating System to deploy on an Nvidia Jetson AGX Xavier as our embedded device. The overall system achieves a high inference speed of 63 frames per second, demonstrating the capability of our system to perform in real-time. Furthermore, we introduce CeyRo, which is the first ever large-scale traffic sign and traffic light detection dataset for the Sri Lankan context. Our dataset …

Recent work done on lane detection has been able to detect lanes accurately in complex scenarios, yet many fail to deliver real-time performance specifically with limited computational resources. In this work, we propose SwiftLane: a simple and light-weight, end-to-end deep learning based framework, coupled with the row-wise classification formulation for fast and efficient lane detection. This framework is supplemented with a false positive suppression algorithm and a curve fitting technique to further increase the accuracy. Our method achieves an inference speed of 411 frames per second, surpassing state-of the-art in terms of speed while achieving comparable results in terms of accuracy on the popular CULane benchmark dataset. In addition, our proposed framework together with TensorRT optimization facilitates real-time lane detection on a Nvidia Jetson AGX Xavier as an embedded system while achieving …

Making a single network effectively address diverse contexts---learning the variations within a dataset or multiple datasets---is an intriguing step towards achieving generalized intelligence. Existing approaches of deepening, widening, and assembling networks are not cost effective in general. In view of this, networks which can allocate resources according to the context of the input and regulate flow of information across the network are effective. In this paper, we present Context-Aware Multipath Network (CAMNet), a multi-path neural network with data-dependant routing between parallel tensors. We show that our model performs as a generalized model capturing variations in individual datasets and multiple different datasets, both simultaneously and sequentially. CAMNet surpasses the performance of classification and pixel-labeling tasks in comparison with the equivalent single-path, multi-path, and deeper single-path networks, considering datasets individually, sequentially, and in combination. The data-dependent routing between tensors in CAMNet enables the model to control the flow of information end-to-end, deciding which resources to be common or domain-specific.

This paper describes a hardware and software codesign approach for object detection and tracking algorithm which is based on modified ViBe background subtraction method and scale adaptable particle filtering algorithm. The modified ViBe performs with a close accuracy to the original work of ViBe but with less complexity and it outperforms the Mixture of Gaussian(MoG) method. Since the modified algorithm has a low off chip memory access requirement, the real time object detection in embedded level with the help of hardware acceleration techniques could be achieved. The implemented particle filtering algorithm is capable of adapting to the scale variation of the tracking objects. After careful investigation of the time consumption profile of each function in our algorithm, the complete algorithm was implemented in IoT edge devices targeted Avnet Ultra 96 development board which is based on Xilinx Zynq …

Fitness related wearables have become ubiquitous in the recent past. Nevertheless, short battery life of these devices is still a pressing issue. Limited battery capacity in small form factor and power hungry continuous monitoring of accelerometer have been significant concerns in this regard. To address these issues we propose a novel low-power step counting solution based on an Electromagnetic energy harvesting mechanism. Extremely simple nature of the step counter removes the requirement of any step detection algorithm, thereby reducing the power consumption, while the energy harvester generates a portion of energy requirement prolonging the battery life.

The aim of this study is to design a Home Automation system by interfacing Power Line Communication (PLC) with GSM communication which would enable the user to control appliances by sending commands via SMS. A communication model based on Star Topology is utilized in this home automation system which can be scaled at a low cost. The communication system facilitates secured communication by using BlumGoldwasser cryptosystem. The communication system employs a master Central Control System placed at the switch board and slave receiver nodes in household appliances. This system was planned from the outset to make as simpler and cost effective as possible so that it would be affordable to more households in low income countries. The system was tested in a lab environment at a maximum distance of 20 m and the operation was verified. Subsequently, Signal levels were obtained at …

This paper describes a demand management system based on narrow band Power line communication and GSM technologies which can be used by the electricity provider to reduce electrical load by turning off nonessential equipment in establishments during high demand periods where enough electricity cannot be supplied. This solution is thought to have less impact on the lives of the public. The system uses Blum-Goldwasser cryptosystem for secure power line communication and to mitigate rogue devices. The system does not require internet connectivity. Therefore, ideal for countries with less internet coverage.

1.1 Background 1 1.2 Problem Statement 3 1.3 Proposed system 4 1.4 Objectives 4 1.5 Deliverables 4

We present a sensor platform to be mounted on the white cane used by the visually impaired. This can estimate the direction of user movement and enable the user to detect obstacles lying on the path in advance. The sensor platform contains an ultrasonic sensor and an IMU (Inertia Measurement Unit). We develop a model to estimate distances to obstacles in the path and their width based on sensor measurements. The model is demonstrated to have an overall accuracy of 84%, with accuracy as high as 90% for obstacles within 50cm in front of the user. Knowledge of obstacle locations and their size in advance, would enable us to guide visually impaired persons to deviate from their path and return to it after the obstacle has been passed.

Tunable laser system requires high resolution and accuracy in position of the wave selective optical element, which can be attained through the use of stepper motor. This paper focuses on the stepper motor controller, with the feedback to improve the accuracy and the microstepping technique to improve the resolution. Sine cosine and trapezoidal current controlling techniques for the micro stepping are used in two different types of situations. The proposed controller uses the linear interpolation and sine cosine current controlling technique to double the rotation accuracy than the resolution of the encoder and detect the obstacle by using the encoder and the static characteristic of the stepper motor. Also, the variable step interval technique is proposed to overcome the speed variation in the trapezoidal current controlling technique. The simulation and the experimental results show that, proposed controller can …

A satellite elephant collar is a device mounted on an elephant's neck to track its position and movements. This information is essential for conservation and management of this highly endangered species. Currently available units are proprietary designs by few manufacturers and no literature can be found about their internal operation. This lack of published information hampers the evolution and development of future collaring systems with increased features and functionalities for research, conservation and management. Critical issues for improvement include live time of the collar, which is related to power consumption and operational modes of the unites, and remote programmability allowing for changes in data acquisition during deployment. Most commercially available collars lack the ability to change data acquisition schedules remotely over the air, and therefore data acquisition is un-optimized, which is …

The growth of the Sri Lankan healthcare sector is a necessity at this stage due to an increasingly ageing community and a significant increase in non-communicable diseases. Despite many regulations that given the sector, it continues to be inundated with many problems in relation to delivery of healthcare services./ Although the state is a prominent player in the industry, private sector healthcare facilities provide curative out patitent care to 60% of the population in the country. The World Health Organisation strategic agenda, Strategic Framework for Health Development in Sri Lanka as well as the Health Master Plan had highlighted the necessity of Information Systems for improved service delivery in the sector./ The existing Information Systems with it's administrative and financial bias has failed to add value in the interaction between the healthcare worker and the patient. The study investigates the possible …

As the problem of urban traffic congestion spreads, there is a pressing need for the introduction of advanced technology and equipment to improve the state of the art of traffic control. In this context, the techniques used for sensing traffic flow information plays a vital role, ensuring accurate real time controllability. Closed Circuit Television (CCTV) has gained popularity because of its ability to provide diverse information on relatively large regions leading to opportunities for performing substantially more complex tasks than conventional detection schemes. By processing these video images, traffic parameters such as speed, traffic composition, queue length etc. can be extracted. In addition CCTV images can be further processed for other useful information such as detection of vehicle shapes, vehicle types, occurrence of traffic violations and vehicle identification numbers etc. To introduce CCTV based vehicle …

Real-time image processing demands much more processing power than a conventional processor can deliver. As a result hardware acceleration became necessary to augments processors with application-specific coprocessors. Due to the limited resources on FPGA and nature of some sequential algorithms, it is difficult to depend entirely on slice resources. In this research, we implemented a multiprocessor architecture to support real-time image processing on FPGA. Furthermore, we benchmarked and compared our implemented architectures with their counterparts. The operational structure of multiprocessor architecture consists of on-chip processors implemented in a parallel manner with efficient memory and bus architectures. The performance properties such as accuracy, throughput and efficiency are measured and presented. Multiprocessor systems are effective in software level parallelism on FPGA. Our …

A single-chip FPGA implementation of a vision core is an efficient way to design fast and compact embedded vision systems from the PCB design level. The scope of the research is to design a novel FPGA-based parallel architecture for embedded vision entirely with on-chip FPGA resources. We designed it by utilizing block-RAMs and IO interfaces on the FPGA. As a result, the system is compact, fast and flexible. We evaluated this architecture for several mid-level neighborhood algorithms using Xilinx Virtex-2 Pro (XC2VP30) FPGA. Our algorithm uses a vision core with a 100 MHz system clock which supports image processing on a low-resolution image of 128×128 pixels up to 200 images per second. The results are accurate. We have compared our results with existing FPGA implementations. The performance of the algorithms could be substantially improved by applying sufficient parallelism.

This paper presents a development of a holonomic mobile robot with remote operation capability. Most of the currently available holonomic mobile robots find problems in field applications because they has not been adopted to field environment.To address those problems we have develop a holonomic platform with a capability of carrying 30 kg payload. This platform use normal car type wheels and a complex multi-processor control system controls each wheels steering and driving to achieve the correct holonomic motion. We have tested it in indoor and out door environments and it's design , hardware and control is presented in this paper.

Designing a PABX system using the existing power line infrastructure as the transmission medium is an effective method of implementing a fast and easily installable communication solution. The scope of this research is to develop such a system for a small scale organizational environment to avoid the complexities and time consumption of cable installation. As the power line transmission technique, we used narrowband frequency shift keying to counter the noise disturbances in power line. In addition we used a novel communication protocol to further reduce the effects of noise and to achieve reliable transmission. We integrated the switching function into the central unit which is the central control element in the system. We were able to achieve voice communication at the rate of 2kbps and to switch between four extensions. Further improvements in the voice quality can be achieved by using advanced noise …

Bluetooth is a high end wireless technology designed for short range communications with high means of integrity. The application of Bluetooth technology for remote monitoring and control is an evolving concept in automation. However, in these applications, the controller which is controlled via Bluetooth technology or the sensor which transmits the sensed values via Bluetooth technology is application-specific. On the other hand, more productivity can be achieved by making the module general purpose, such that it can be used for wide range of applications. In fact, the general purpose concept can be applied to a selected set of devices with standard data interfaces, by making the system general purpose to the developer rather than the end user. In this work, a general purpose controlling module is designed with the capability of controlling and sensing up to five devices simultaneously. The communication …

This paper discusses the design of a centralized control and monitoring system for railway traffic that can be implemented on top of the existing control logic (interlocking system) of Sri Lanka railways. The overall solution comprises of a central server system and databases with redundancy, relay house hardware unit (with embedded Ethernet and a custom communication protocol for microcontrollers) to interface with the relays of the interlocking system, Dispatchers Controlling Interface for remote monitoring and controlling, and a TCP/IP network which interconnects all the above mentioned components of the total system. The current centralized traffic control (CTC) system in Sri Lanka, located at Maradana, was implemented in the year 1985. This has manually operated switch boards, a relay based operation (separate relays for controls and indications) having every switch hardwired to the relays which are …

In this paper we are presenting a low cost, locally developed customized monitoring and controlling system for a water purification plant. This system can facilitate monitoring and controlling of the entire plant from a single location. Apart from the major functions of monitoring and controlling through a GUI, additional functionalities such as providing remote terminals to the management via a LAN, generating reports, keeping logs in historical database are provided through the system.