Technical scheme of safety monitoring system for urban rail transit operating environment

As a green and environmentally-friendly transportation method, urban rail transit has made great progress in recent years. Especially during the "13th Five-Year Plan" period, the mileage and coverage of new urban rail transit lines have increased dramatically. While urban rail transit brings convenience, safety and comfort have also become two important factors that need to be explored. The paper adopts the experimental analysis method and comparative analysis method, sets up the experimental group and the reference group, uses the Sperling evaluation index, and basically realizes the technical scheme design of the urban rail transit environment safety monitoring system based on smart phones. The research results show that the true value of the displacement in the static experiment of the train is 0, and from the average point of view, the error between the displacement monitoring method used in this experiment and the true value is only 0.0047 mm; from the observation of the standard deviation, the fluctuation of the monitoring data The range is also smaller than that of the laser displacement sensor, which is more stable. The image displacement monitoring method used in the displacement loading test is 0.107 mm, Median filtering is to sort the gray values of other pixel samples in the pixel neighborhood of the image, and take the median of the gray values as the new gray value of the pixel after sorting. Gaussian filtering is a linear smoothing filtering. Its principle is to perform a weighted average of the pixel values in the neighborhood of the target pixel and then replace the original pixel values to eliminate Gaussian noise. Morphological filtering uses mathematical morphological methods for grayscale images and binary images. The most important ones are image expansion and erosion operations, as well as opening and closing operations. And the accuracy is 45% higher than that of the laser displacement sensor. The linear regression equation for the data is y = x − 0.21 and the square of the correlation coefficient is 0.998, which is a high degree of fitting. In the dark environment experiment, the output noise is reduced by the drum noise reduction, and it is basically undistorted. It can be compared with the experimental results of the non-dark working condition at a distance of 10 m. The average value of the two is 0.0029 and 0.0055 basically the same, and the accuracy is 0.13 And 0.06, which is very good. It shows that whether it is a static experiment or a displacement loading experiment, the scheme designed in this article can be perfectly applied, and it is also very suitable for dark conditions (such as subways).