Cloud-precipitation particle image artifact identification and rejecting in airborne optical array probe measurement process

Abstract

The present invention discloses a cloud-precipitation particle image artifact identification and rejecting method in an airborne optical array probe measurement process, belonging to the technical field of image signal processing. The method comprehensively utilizes an arrival time interval algorithm and an isolated particle number threshold algorithm to perform identification of crushing particles broken by mechanical impact of particles and a probe detection arm or turbulence and wind shear interaction of the particles and the housing of the probe in the measurement process, and employs an image feature factor determination method to perform identification of fake particles such as stripe-shaped particles generated by mismatching of a sampling speed and an air speed, and linear particlesand blank particles generated by electromagnetic signal interference. The cloud-precipitation particle image artifact identification and rejecting in the airborne optical array probe measurement process can effectively perform identification and rejection of cloud-precipitation particle images in the airborne optical array probe measurement process and has an important meanings for research of the cloud-precipitation physical process by employing the cloud-precipitation particle image data through airborne measurement.

Description

technical field [0001] The invention belongs to the technical field of image signal processing, and in particular relates to a method for identifying and removing artifacts in cloud and precipitation particle images during the measurement process of an airborne light array probe. Background technique [0002] Using airborne optical array probes to observe aircraft through clouds is one of the main detection methods for modern cloud microphysics research. The data measured by airborne optical array probes can not only be used for cloud precipitation physics research, but also for the verification of satellite radar and other remote sensing results. . Judging from the particle images acquired by the airborne light array probe, the particle images contain many pseudo-particles. These include broken particle artifacts caused by mechanical collisions between the particles and the probe arm during the measurement, or turbulence and wind shear interactions with the probe housing; ...

AI Technical Summary

Problems solved by technology

[0003] The identification and removal of cloud precipitation particle image artifacts in the measurement process of airborne light array probes have been carried out abroad. In 1978, Cooper of the U.S. Bureau of Land Reclamation proposed to use the space distance to filter out the scattered and broken particles. , although the spatial distribution of the broken particles is relatively close, but because the sampling rate of the particle image is proportional to the speed of the aircraft, the spatial distance between the particle images is also relatively close when the aircraft speed is relatively fast, while the actual aircraft flight The velocity of is difficult to keep constant, so the method based on spatial distance has great limitations in filtering out broken particles

In 2005, Korolev of the Canadian Ministry of Environment proposed a method of using the isolated particle number threshold to distinguish particle breakage artifacts. This method assumes that there can only be one isolated particle in a particle image frame, and when a particle breaks, many broken particles are simultaneously recorded in the In the same particle image frame, the isolated particle number threshold can be used to identify and eliminate, but in fact, many broken particle fragments are recorded by the instrument in a single form, so the method based on the isolated particle number threshold is important in the identification of broken particles not high in accuracy

In 2006, Field et al. of the US National Center for Atmospheric Research proposed a method to filter out broken particles using the arrival time interval threshold. The time interval between two natural particles entering the sampling area of ​​the instrument is less than the time interval between two adjacent fragments entering the sampling area of ​​the instrument, and the broken pseudo-particles can be identified with high accuracy, but this method can only be used for the fragments formed after the natural particles are broken Pseudo-particles are effective in the form of more than two consecutive image frames recorded. For the case where all fragmented pseudo-particles are recorded in the form of one image frame, this method cannot identify

Since the generation of image artifacts is not only related to the instrument, but also related to the airborne platform and the measured atmospheric environment factors, the methods currently published abroad cannot be directly applied to the images of the cloud precipitation particle process measured by domestic airborne light array probes data processing

Images