An image-based night vehicle detection method

Abstract

The invention provides a night vehicle detection method based on an image. The detection method comprises the following steps: a computer reads the obtained image, and pretreatment such as grey level transformation, median filtering and morphological processing is carried out on the image; subtraction operation is carried out on the median filtering in the process and the image after being eroded through morphology; binaryzation is carried out on the image, a threshold value is one half of the sum of the maximum and the minimum of a grey value, pixel points larger than the threshold value are initial information of car lamps, expansion processing is carried out on the image for better visual effect, and a part which does not belong to the car lamps exists in the image; the image part of the car lamps of the same car is extracted, and the redundant part is removed; the image output by the computer is the detection result of the car lamp information of headlamps of night vehicles. The method is suitable for night vehicle detection of scenes with fixed cameras, and has the advantages of being high in universality and accuracy and high in reaction speed.

Description

technical field [0001] The invention relates to a computer vision signal acquisition technology, in particular to a method for vehicle detection by collecting nighttime vehicle headlight signals through a computer. Background technique [0002] With the rapid development of social economy and the advancement of modern technology, automobiles have become the most important means of transportation in modern society. To a certain extent, motorization has brought great convenience to people's travel, and at the same time accelerated the pace of social life and improved people's work efficiency. However, while motorization has brought convenience and enjoyment to human beings, it has also made people have to Faced with many problems. With the increase in the number of cars, serious traffic problems have emerged. Traffic congestion can easily cause a large number of traffic accidents to occur frequently. Therefore, how to effectively solve traffic problems has become a common co...

AI Technical Summary

Problems solved by technology

However, in practical applications, the background is often easily affected by factors such as illumination changes, noise, and foreign intrusion due to different shooting times. The background needs to be updated in real time, and the target that is close to the gray level of the background cannot be fully detected.

At this time, if you want to obtain accurate target information, you need to go through a complex algorithm, but the complex algorithm increases the amount of calculation, and the calculation time is long, making the rapidity worse.

However, if you want to improve the speed, you must reduce the amount of calculation, and the reduction in the amount of calculation will make it difficult to guarantee the accuracy

Images