Luminous flux control method of high definition vehicle snapshot system

Abstract

The invention provides a luminous flux control method of a high definition vehicle snapshot system, belonging to the image processing field. The control method comprises the following steps: initializing the parameters of a camera, setting an initial photometric range and a photometric luminance reference value table; obtaining the current shutter speed value, gain value, reference datum value and current photometric interval of the camera; determining the photometric range of an acquired image; calculating photometric luminance value in the designated area of the photometric range; calculating the luminous flux control parameter; sending a control command, waiting in the designated photometric interval, and performing the next adjustment process. The method automatically adjusts the shutter and gain of the acquisition equipment and adjusts the light complement equipment according to the photometric result of a special area so that the pictures shot at any moment have proper exposure luminance, the problem that the exposure effects of the plate number and portraits in a vehicle on the traditional image are not consistent is solved; and the control method can be realized in the embedded system, thus the combination control of multiple devices becomes easy.

Description

technical field [0001] The invention belongs to the field of image processing, and in particular relates to a high-definition vehicle capture system in the field of intelligent transportation, which is suitable for vehicle capture in various complex environments such as highways and city bayonets. Background technique [0002] With the continuous development of intelligent transportation technology, the application of vehicle capture system is becoming more and more extensive. The current mainstream vehicle capture systems usually have the following problems: (1) Two images, the panorama and the feature map, are required to fully express the vehicle information, and the situation inside the vehicle cannot be clearly captured. (2) The feature map only considers the brightness of the license plate, which makes it impossible to see the shape and color of the car body at night or when the light is insufficient. After the supplementary light is enhanced, the license plate may not...

AI Technical Summary

Problems solved by technology

For the current mainstream vehicle capture system, there are usually the following problems: (1) Two images, the panorama and the feature map, are required to fully express the vehicle information, and the situation inside the vehicle cannot be clearly captured

(2) The feature map only considers the brightness of the license plate, which makes it impossible to see the shape and color of the car body at night or when the light is insufficient. After the supplementary light is enhanced, the license plate may not be recognized due to overexposure

(3) The system cannot adapt well to complex road conditions such as highways or expressways, and the obtained image is brighter or darker, and even in cloudy weather, it may even appear flickering

(4) Poor luminous flux control method leads to poor image quality, such as too large gain leads to image noise; too slow shutter makes it impossible to clearly capture high-speed vehicles, etc.

(5) When the backlight is strong, the background is very bright but the license plate and the car body are very dark, making the license plate unrecognizable

The existing technology usually divides the license plate and human face into two images to capture, and exposes them to different degrees respectively. There are three problems: (1) a large amount of storage space or higher transmission bandwidth is required

(2) The host needs to have a high processing speed

(3) The calculation is complex and cannot be applied to embedded systems. The light control process requires the participation of a PC, and the joint control of multiple devices is complex

Images