Adaptive object classification device having parallel framework and method

Abstract

The invention discloses an adaptive object classification device having a parallel framework and a method. The adaptive object classification device having a parallel framework includes a step of storing multiple scene parameters and multiple classifier parameters, first at least one piece of image data is captured, a plurality of obstacle images in the image data are selected, and an appropriate selection range is selected according to each obstacle image range. Then a plurality of image processing units are utilized to correspondingly calculate multiple pieces of obstacle characteristic data of the plurality of obstacle images in a one-to-one parallel processing manner, the corresponding scene parameters and corresponding plurality of classifier parameters are selected according to the multiple pieces of obstacle characteristic data to calculate, thereby obtaining multiple pieces of classification data, and if the classification data are judged to be a preset obstacle category, positions of the selected obstacle images are output, and thus the purpose of detecting obstacles in real time can be achieved.

Description

technical field [0001] The present invention relates to an adaptive object classification device and its method, in particular to an adaptive object classification device with parallel architecture and its method. Background technique [0002] As driving safety has been paid more and more attention in recent years, with the sharp reduction in the cost of imaging equipment and the maturity of image recognition technology, the application of image recognition systems in vehicle safety has become more and more extensive, and in the security system, the use of image recognition is to reduce the overall system An important method of cost. According to statistics, drivers can avoid at least 60% of rear-end collisions, 30% of head-on collisions, and 50% of road-related accidents as long as they get an early warning 0.5 seconds before the collision. 90% of accidents can be avoided if the warning time is shortened. However, in the image recognition system, the huge amount of calcul...

AI Technical Summary

Problems solved by technology

[0006] Another object of the present invention is to provide an adaptive object classification device with a parallel architecture and its method, which can determine the corresponding scene according to the captured image, and can flexibly adjust the classifier parameters according to various scenes, It can calculate more accurate obstacle images and their types to solve the problem of misjudgment by existing detection systems

Images