Avionic image stabilizer

Abstract

The invention discloses an avionic image stabilizer which is based on a TMS320DM642 chip to build a high-performance digital media processing platform, wherein an avionic image stabilization system simultaneously integrates a CPLD chip to manage the chip logic in the system, a two-way PAL video input port and a one-way PAL video output port, an SDRAM and an FLASH memory as well as an Ethernet interface and an RS485 serial drive. In addition, the invention further relates to a video processing method, stable video image output is realized by video input and storage, image processing, video output and TV output. The invention not only distinguishes the difference and relation of overall situation movement, main movement and random jitter, particularly in the forecast period of the main movement, a secondary filter method is utilized to effectively distinguish the main movement and the random jitter, by the implementation on hardware, good effect is obtained, the video image is stable after being outputted by the image stabilization system, and the image stabilization effect is achieved.

Description

technical field [0001] The invention belongs to the field of image acquisition and processing, and relates to an avionics image stabilizer, more specifically, an electronic image stabilization system that can be used in the fields of ground observation and reconnaissance of aerial moving carriers. Background technique [0002] As an important source of information, image information makes electronic camera systems more and more widely used in various carriers, especially in the new earth observation system with long battery life, real-time image download and operation in high-risk areas—aerial remote sensing In the system, the application of camera system makes aerial remote sensing play an important role in disaster monitoring, environmental monitoring and security monitoring. However, the effect of the video image obtained by the camera system in actual operation is affected by the attitude changes or vibrations of the carrier at different times, which is reflected on the ...

AI Technical Summary

Problems solved by technology

[0006] Since the inter-frame global motion vector of aerial video images is composed of active motion vectors and random shaking motion vectors, Literature 1 only has global motion estimation, without the separation of main motion and random shaking. Using the obtained global motion vectors for compensation cannot produce very large Good image stabilization effect, document 2, although the Kalman filter method is used in the main motion prediction, for the video stabilization system, the constant speed camera model is usually used to establish the system equation, and the estimated global inter-frame motion vector is accumulated to obtain the absolute frame Displacements, as observations from the observation equation

In practical video stabilization systems, the size of the compensation vector is limited by the system, and the results of Kalman filtering often cannot be directly applied to video compensation

Images