Study on Illumination Measurement Method of Lighting Environment Based on RBF Neural Network

Abstract

In order to better measure the indoor light environment, evaluate the quality of the indoor lighting environment, and improve the requirements of lighting comfort, a camera image measurement illuminance method based on RBF(Radial basis function)neural network is proposed, and the camera sensor imaging theory is derived and analyzed to obtain the environmental illuminance and camera sensor The relationship between the parameters, the establishment of the RBF neural network model, by building an experimental system platform, collecting data sets to train the network model, fitting the neural network model parameters to obtain the illuminance measurement model, and use the image gray level and the reference point illuminance as the neural network Input and ambient illuminance as output. The prediction result shows that the error between the illuminance value predicted by the neural network and the actual measured value of the illuminance meter is within 10lx, and the relative error is less than 8%, which meets the requirements of lighting building design standards. Therefore, this method can achieve rapid measurement of the illuminance in the environment, and has a relatively high High precision.