Automated Extraction Of Street Lights From JL1-3B Nighttime Light Data And Assessment Of Their Solar Energy Potential

The production,living and development of cities are inseparable from a large amount of energy supply.According to the statistics of the National Energy Administration,the annual electricity consumption of urban lighting accounts for 4%to 5%of the total annual electricity consumption in China.As the main equipment of urban road lighting,street lamps are mostly high-pressure sodium lamps with high energy consumption.At present,about 60%of China’s power generation is still provided by fossil energy-consuming thermal power generation,and the consumption of non-renewable fossil energy has caused serious environmental problems.Solar energy is a renewable and clean energy.At present,some cities have begun to use solar street lamps to replace traditional street lamps to alleviate resource loss and secondary environmental problems.To use solar street lights reasonably and effectively,one must know the location of the street lights and their solar energy utilization potential.Traditional street lamp position extraction algorithms mostly rely on on-board LiDAR data or manual on-site investigation.However,due to the high cost of LiDAR data and the high cost of on-site investigations,it is difficult to use traditional methods in large-scale scenes,and it is difficult to quickly identify the type of street lights.The Jilin No.1 video 03 star（JL1-3B）luminous remote sensing image is the next generation of new luminous data with three bands of information and high spatial resolution,which provides the possibility of extracting the location of street lamps and identifying the types of street lamps on a large scale.Therefore,based on JL1-3B luminous remote sensing data,this paper proposes a new method for street lamp position extraction and street lamp type discrimination,and estimates the solar energy utilization potential of the street lamp location by combining digital surface model data（DSM）.The main research content and results include:The innovation of this paper is mainly based on multi-source remote sensing data.A new method of street lamp position extraction and street lamp type differentiation is proposed,and the utilization potential of solar street lamps in the study area is estimated.Moreover,this method can be quickly and accurately applied to street light extraction and solar street light utilization potential assessment in a large area and the entire city.The main steps and corresponding results of this method are divided into the following three parts:（1）This article first used a skylight meter to conduct a field survey on the radiance of the roads in the study area.By analyzing its spatial characteristics,a method of using local maximums to identify the location of street lights was determined;open street map（OSM）and specific The range of the buffer is used to extract the road area from the JL1-3B luminous image as the limited range for extracting street lights;then use the difference in the performance of the lights and street lights in the background pixels in the image and the average filter to remove the interference of the lights and building decoration lights At the same time,it also solves the error of extracting multiple street lamp points caused by multiple local maximum values within the same street lamp range;finally,the local maximum method is used to solve the radiance value obtained by synthesizing the three bands of JL1-3B,To obtain the location of the street lamp in the study area,the extraction accuracy is about 90%;based on the maximum likelihood classification method,using the three band information of JL1-3B,successfully identify high-pressure sodium lamp（HPS）street lamps and light-emitting diode（LED）street lamps,The overall classification accuracy is around 99%.（2）This article will input DSM data and hourly cloud cover data into the SHORTWAVE-C solar radiation model,and calculate the direct solar,scattered and reflected radiation of each grid position in the 2016 study area at 10-minute intervals The annual average solar radiation values of all grid positions in the study area were obtained.Combined with the street lamp position data obtained before,the accurate average annual solar radiation value of each street lamp in the study area can be estimated.In the study area,the minimum annual solar radiation value of the street lamp is 2.64 MJ/m²/day,and the maximum annual solar energy The radiation value is 22.96MJ/m²/day.（3）According to the electric energy required by solar street lamps,the amount of solar radiation received,the photoelectric conversion efficiency and the attenuation coefficient,the area of solar panels required for various types of street lamps can be determined.Take the maximum value of the area of each solar panel as the recommended area for the refined installation of solar street lamps.By simulating the use of solar street lamps to replace the original high-pressure sodium lamp,it was found that during its 20-year service life,it can save 1.85×10⁴ KWH of electrical energy,7.41 t of standard coal,5.03 t of C emissions,18.47 t of CO₂ emissions,0.55 t of SO₂emissions,and 0.28 t of NO_X emissions.In summary,the method of street lamp recognition and solar street lamp utilization potential assessment based on multi-source remote sensing data proposed in this paper can be quickly and accurately applied to the refined installation and potential assessment of urban solar street lamps,and make relevant decisions for relevant government departments Provide scientific advice and basis.At the same time,since most of the data used in this article is remote sensing data,it is highly available and generalizable.