In this paper, among the W-S-R relationship methods proposed by Lee, et. al., (2020) to produce rain-based rain information in real time, we tried to produce actual rain information by applying machine learning techniques to take into account the effects of wiper operation. To this end, rain sensor proposed the Graded Descent and Threshold Map method for pre-processing the cumulative value of the difference before and after wiper operation by utilizing four sensitive channels for optical sensors developed by Kim Byung-sik (2016) and using rain sensor data produced by five rain conditions in indoor artificial rainfall experiments. This method is the method of producing rainfall information by calculating the average value of the Threshold according to the rainfall conditions and channels, creating a Threshold Map corresponding to the grid 4 (channel) x 5 (thinking of rainfall information) and applying Optima Rainfall Intensity among the big data processing techniques. For the verification of these proposed findings, the application was evaluated by comparing the rainfall observations with the methods presented by Lee, et. al., (2020).