基于深度机器学习的霾污染监测技术

传统的霾污染监测技术监测准确率低，收集的图像完整度差，为了解决上述问题，基于深度学习研究了一种新的霾污染监测技术。通过污染数据收集精准划分其产生的地点，整合获取的追踪信息，在三维分布空间图掌控霾污染可能存在的条件，多次进行机器飞行追踪实验，根据不同的污染项目组对霾污染进行数据监测，根据霾污染数据的浓度信息以及深度机器学习的输入数据类型对收集数据进行分类，查询数据类型，同时监测气溶胶的厚度、霾污染中具有毒性的二氧化硫及二氧化氮物质以及兴趣区域。为验证技术的有效性，设定对比实验，结果表明，基于深度机器学习的霾污染监测技术监测结果准确率为90%，图像收集完整度平均值为82%，具有更强的监测能力。

Haze pollution monitoring technology based on deep machine learning

Traditional haze pollution monitoring technology has low monitoring accuracy and poor integrity of collected images. In order to solve the above problems, a new haze pollution monitoring technology is researched based on deep learning. Accurately divide the place where it is generated through pollution data collection, integrate the acquired tracking information, control the possible conditions of haze pollution in the three-dimensional distribution space map, conduct machine flight tracking experiments many times, and monitor the haze pollution data according to different pollution project groups According to the concentration information of haze pollution data and the input data type of deep machine learning, the collected data is classified, and the data type is queried. At the same time, the thickness of the aerosol, the toxic sulfur dioxide and nitrogen dioxide substances in the haze pollution, and the area of interest are monitored. In order to verify the effectiveness of the technology, a comparative experiment is set up. The results show that the accuracy of the monitoring results of the haze pollution monitoring technology based on deep machine learning is 90%, and the average of the completeness of image collection is 82%, which has stronger monitoring capabilities.