Streetlight that emits too much light or shines when and where it‟s not needed is wasteful. Wasting energy has huge economic and environmental consequences. As well as due to the manual nature of operation, streetlights are most often seen turned „on‟ in the daytime and turned „off‟ during evening time in Bangladesh that is pretty unfortunate to observe even in this age of twenty first century. Automated streetlight management system is needed to resolve these problems. This study aims to develop a novel streetlight management system powered by computer vision technology mounted with the CCTV camera that allows the LED streetlight to automatically light up with proper brightness by recognizing the presence of pedestrians or vehicles and reversely dimming the streetlight in their absence by semantic image segmentation from video. Subsequently, our model can recognize the daytime and nighttime by itself that made the automation process of turning „on‟ and „off‟ the streetlight possible to minimize the cost of energy consumption. We trained the U-net model based with ResNet-34 as backbone to accomplish the tasks. The validity of the models is ensured through corresponding evaluation matrices. The proposed model is much simpler, cost effective, durable and robust than the conventional, physical sensor-based existing intelligent solutions.

Energy Efficient Automatic Streetlight Controlling System using Semantic Segmentation

One of the world's most public serious issues is road fatalities. Every year, millions of people die or suffer non-fatal injuries or impairments in car accidents worldwide. Using helmets, wearing seat belts, avoiding drunk driving, and adhering to traffic laws can all help to reduce fatalities and injuries. Road accidents are caused by driver's poor decisions or actions while driving. Furthermore, accidents can occur as a result of vehicle component failures or the drivers' health. Controlling speed and monitoring the health of the driver/passengers can help to improve vehicle and passenger safety. Automobiles may be equipped with driver/passenger health monitoring systems with sensors to help prevent accidents. The proposed work suggests a sensor-based prototype for autonomous vehicles with smart seat belts, and road rules such as speed control, stopping the vehicle when the traffic signal shows red light, parking/no parking control, and horn/no horn control systems using Radio Frequency (RF) communication. The prototype developed works in both manual and autonomous modes. Furthermore, this article also suggests including the YOLOR (You Only Learn One Representation) for real-time obstacle detection for self-driving and autonomous cars, as YOLOR demonstrates 5 percent higher performance than YOLOv4 (You Only Look Once version 4). 

Accident Prevention and Emergency System for Vehicles and Passengers Safety

One of the most neglected sources of energy loss is streetlights which generate too much light in areas where it is not required. Energy waste has enormous economic and environmental effects. In addition, due to the conventional manual nature of the operation, streetlights are frequently seen being turned ON during the day and OFF in the evening, which is regrettable even in the twenty-first century. These issues require automated streetlight control in order to be resolved. This study aims to develop a novel streetlight controlling method by combining a smart transport monitoring system powered by computer vision technology with a closed circuit television (CCTV) camera that allows the light-emitting diode (LED) streetlight to automatically light up with the appropriate brightness by detecting the presence of pedestrians or vehicles and dimming the streetlight in their absence using semantic image segmentation from the CCTV video streaming. Consequently, our model distinguishes daylight and nighttime, which made it feasible to automate the process of turning the streetlight 'ON' and 'OFF' to save energy consumption costs. According to the aforementioned approach, geolocation sensor data could be utilized to make more informed streetlight management decisions. To complete the tasks, we consider training the U-net model with ResNet-34 as its backbone. The validity of the models is guaranteed with the use of assessment matrices. The suggested concept is straightforward, economical, energy-efficient, long-lasting, and more resilient than conventional alternatives.

CNN based Intelligent Streetlight Management Using Smart CCTV Camera and Semantic Segmentation

Conventional public streetlight controlling and monitoring systems are carried out locally without being able to be carried out remotely. Hence, it is slow to handle if there is damage. Regarding security, it is not uncommon for the public streetlight to become the object of vandalism and theft. The existence of smart public streetlights with IoT technology that can display the current status of public streetlights is urgent. In this study, our main intention is to develop a smart streetlight monitoring system that has exclusive features by integrating various devices. All data generated from integrated devices will be displayed in real-time via the IoT-Blynk platform and recorded on a Google Spreadsheet per minute. A smart streetlight system monitoring is implemented on the main road of the Jakarta State Polytechnic to determine the system’s performance. This system has been functioning correctly, as evidenced by the reading and sending of all sensor data appropriately to smartphones and Google Spreadsheets via a Wi-Fi network connection. Based on data analysis, the total energy consumption required of smart streetlights for 24 hours is 0.117 kWh. 

A Real-time Performance Monitoring of IoT based on Integrated Smart Streetlight

In the modern world, an increase in the usage of automobiles for commercial purposes has also increased the number of accidents occurring in commercial vehicles, which leads to the loss of life of the people involved in the accident. To minimize the death rates involved in an accident, the people who are met with the accident must claim medical assistance at the correct time. This study is concerned with two set-ups. One set-up is associated with the vehicle, where the use of a MEMS or gyroscopic sensor, a vibration sensor, and a gas sensor integrated with Arduino helps to detect the accident. Here, the location is detected by the GPS module and updated in the cloud by using the ESP8266 Wi-Fi module. If any accident is detected, the RF transmitter circuit sends the signal to the RF receiver. The other configuration is related to the Ambulance which consists of an RF receiver circuit integrated with the NodeMCU microcontroller. When the signal reaches the receiver, NodeMCU retrieves the information from the cloud and displays it on the LCD. Integration of a tracking system with a Radio frequency transmitter and receiver helps build IoT services using embedded systems. The system of providing medical assistance to the people involved in the accident would help us reduce the death rates.

Design and Implementation of IoT Based Accident Detection and Prevention System

At present, road accident has become a regular headache for all people around the world. To overcome the situation, people are thinking of updating the way of driving and road law. Different brands of automobile companies are trying to invent different types of facilities in vehicles. The main focus of this paper is to ensure the life security of the drivers. There will be sensors for health monitoring system that will track the driver's health, such as heart rate, temperature, and whether or not the driver is drunk, as well as drowsiness. This prototype uses an ultrasonic sensor, which will detect very close vehicles and reduce its own speed. However, the engine will start based on driver health, where sensors will always observe the driver's lively movement. Car accident prevention with the health monitoring system of the driver can be a blueprint to reduce car accidents, and it will be cost-effective. This project is expected to have a significant effect on society and obtain better results.

Car Accident Prevention And Health Monitoring System For Drivers

IoT-Based Smart Street Light for Improved Road Safety

Technology’s extensive use in the twenty-first century has a negative impact on the environment. Researchers are making more progress on renewable technologies to mitigate global warming, while demand for advanced technology is also gaining greater interest for safety purposes. Traditionally, most high-energy consumption street lights have been operated manually. Also, road accidents cannot be easily identified in real-time due to a lack of immediate data to the authorities. Numerous solutions have been taken to solve these problems but none of them ensures an optimum solution. Internet of Things and AI-based integrated control system can be implemented to solve this problem that can monitor street light intensity, identify car accidents with name-plate detection, and identify the face in unusual situations when someone presses an emergency push button. This automated system will ensure protection and power consumption in this way. This paper will illustrate the existing automated system and the suggested system.

IoT Based Smart Self Power Generating Street Light and Road Safety System Design: A Review

Extensive research is taking place around the world and new technologies are emerging that aim to make people's lives better and easier. Energy can be seen as an essential element for the development of society. People use electricity for a variety of reasons, and the use of electricity is growing by the day. In addition, the use of electricity on a regular basis must be regulated and managed as necessary. In remote island, maximum electricity gets out from using fossil fuel. It is full of renewable energy sources due to the geographic conditions to generate electricity instead of using fossil fuels. The electricity crisis has become a major constraint on continued economic growth, and the remote island energy sector is currently facing serious challenges. Due to the depletion of nature and the energy policy dependent on single fuels, the search for an alternative energy source has become urgent for the country. This study examined how to overcome such a scenario using alternative energy sources such as solar, wind, and hydroelectricity. This study was conducted to assess the optimal feasibility of a grid-connected renewable electricity system in a coastal region of remote island. The design and simulation are operated by the Optimized Mixture of Multiple Energy Sources (HOMER) software to obtain the most economical, the most energy-balanced and the most ecological solution by comparing each model of optimal shape. The HOMER simulation was performed in terms of lowest energy cost, current net cost, short payback period, highest renewable fraction.

Renewable Power System Calculation and Management for Remote Island

In a progressing country like Bangladesh, having one of the largest populations in the world and where almost 50% of Bangladeshi people are fundamentally employed in agriculture. At the same time, whenever calculating the amount of food loss in the world, we can see that there are about 500 types of planted grains, fruits, and vegetables. If all these food items can be stored then it will be observable that the amount of food waste will be reduced. Preharvest and storage forfeiture is a vital plight, that needs to be consigned in careful examination and inspection. There are many foods conservation operating procedures like cold storage, drying, and many more that have been developing for many years to tackle the upward of losses. There are many methods for food preservation that are sustainable and need hours. In this state, solar drying the one of the best. In most countries, distinct types of solar dryers are developing and at the same time, a good portion of the research is continuing for the value inclusion of agriculture products by generating solar dryer technologies. This paper will show the different classifications of drying, some drying criteria, and more information about this drying method. The main focus is this paper is to develop an environmentally sound solar system hybrid dehydration. This prototype will present and develop an easy control for farming sectors. There is a device to collect voltage drop and current flow and make performance scanning of the solar system hybrid dehydration. It will determine the battery pack cost depending on battery pattern and will also recharge the battery lifetime. It will also decrease the total cost and focus on the quality of the products. A hybrid dehydrator for foods continuation by using a solar system will be cost-effective and can be an outline to minimize food wastage. This project is anticipated to have a noteworthy consequence on society and acquire finer results.

Touhidul Islam Talukder

Papers

Car Accident Prevention And Health Monitoring System For Drivers

IoT-Based Smart Street Light for Improved Road Safety

IoT Based Smart Self Power Generating Street Light and Road Safety System Design: A Review

Renewable Power System Calculation and Management for Remote Island

Hybrid Dehydrator for Foods Preservation by Using Solar System