基于对照区设置的结化霜曲线去除雾噪声的方法实现

摘要

The object of this dissertation is the defrosting video of single air-conditioner.Based on the changes of image gray,trost-defrost process time series data is obtained.Then,fogging noise data smoothing is mainly studied.A method of setting synchronous sampling contrast area is proposed.Contrast region is used to monitor fog's appearance and dissipation,and then fog existence intervals are recognized and oriented.Based on linear interpolation method,inner data within the relevant intervals of frost curve can be adjusted through interpolated reconstruction.By above means,the fog noise of process data can be removed,and data without fog noise can be unaffected.At last,some applicable conditions are provided.Experiments show that the algorithms based on contrast area setting can obtain good curve smoothing results.The principle of this method is simple,and it has a good effect of filtering and smoothing of the data curve.In the field of curve smoothing,it has some application space.%以空调结霜、化霜的视频为研究对象,基于画面灰度特征变化,获取结化霜过程的时序监控数据;然后,重点研究了含雾气噪声时序数据的平滑问题;提出一种通过设置同步采样的对照区的方法,监测雾气出现与消退的动态变化,用以识别定位雾气的存在区间;再基于线性内插值方法,对结化霜时序曲线的相关区间数据进行差值重构,实现滤除雾气噪声同时不影响无噪声数据质量的目标;最后给出本方法适用的一些条件;实验表明,基于对照区设置的方法能够对雾气干扰取得很好的曲线平滑效果;且这个方法的实现原理简单,对数据曲线的消噪平滑效果比较好,在曲线平滑应用领域可以有一定的应用空间.