Low-Light Image Enhancement for Dashcams

Image acquisition system that uses an image sensor with both visible light and infrared sensors to capture better images in low light conditions. The system determines the exposure time for the visible light sensor based on the ambient light level. This improves the quality of the visible light image. It also captures an infrared image using the infrared sensor. The infrared image is processed to generate a wide dynamic range (WDR) image. The visible light image and the WDR image are then fused to create a higher quality final image compared to just using the visible light image alone in low light conditions.

Source

A vehicle vision system that enhances night vision for a vehicle driver. The system uses two cameras facing forward - one with a wide field of view for machine vision applications, and one with a narrower field of view optimized for low light color capture. The narrow FOV camera has optics and spectral filters to provide enhanced color imaging even in low lighting conditions like night driving. The narrow FOV camera captures color video which is displayed to the driver for enhanced night vision, while the wide FOV camera is used for machine vision functions like object detection.

Source

Imaging device with improved low-light performance by combining visible and infrared images. The device has a visible light imaging element and an infrared light imaging element with different exposure rates. It combines the visible light image with the fast-rate infrared image and the slow-rate infrared image. This increases dynamic range without reducing frame rate compared to just increasing infrared exposure. The device also optimizes exposure for each element to prevent charge accumulation during high infrared light levels.

Source

Enhancing image quality from a vehicle's camera to enable autonomous driving in challenging lighting conditions. The method involves proactively and reactively adjusting a light-control feature on the vehicle to mitigate issues like blooming, glare, and lens flare when external light sources like the sun are present. If the image quality is expected to be low due to light conditions, the feature is adjusted to reduce the amount of light encountered by the camera. If a specific light source like a traffic signal is detected, the feature is adjusted to block that light. This improves image quality and allows accurate object detection for autonomous driving even in difficult lighting scenarios.

Source

Optimizing camera images in vehicles by selectively binning pixels based on light conditions. The technique involves determining light patterns around the vehicle using sensors, location data, etc. Then for each camera image, binning pixels together in dark regions to improve handling and viewing, but leaving pixels unbinned in bright regions for better resolution. This results in frames with binned and unbinned regions covering both light conditions.

Source

Image processing device for vehicle cameras that improves object recognition in low beam headlight conditions by adjusting pixel values. The device acquires images from the vehicle's camera, determines if the headlights are in low beam mode, and then adjusts the relationship between object luminance and pixel values to compensate for the low light conditions. This raises pixel values in areas of the image with low luminance objects to improve recognition.

Source

Camera system for vehicles that adaptively applies image correction based on scene brightness to compensate for lens vignetting without reducing frame rate. The camera activates a lens correction function to compensate for light falloff in the image edges caused by wide-angle lenses. But it only does so when the scene brightness is low. This prevents false brightness interpretation due to dark current, which would normally reduce frame rate. The camera captures scene brightness and uses it to determine when to enable lens correction. This allows compensating for lens vignetting without sacrificing frame rate in low light conditions.

Source

Adaptive histogram spreading for cameras in vehicles to improve image quality in low light conditions while preventing over-darkening of images. The method involves setting limits on the output pixel values based on the environment brightness. This prevents further darkening of already dark pixels when histogram spreading is applied in low light. The limits are determined by acquiring a parameter related to the environment brightness, like the sensor value from a dedicated brightness sensor. By dynamically limiting the output pixel values based on the environment brightness, the camera avoids over-darkening in low light while still providing improved contrast compared to the raw image.

Source

Reducing motion blur in low light conditions by optimizing exposure time based on image brightness levels and distribution. The camera changes exposure time based on the brightness level and brightness distribution in an image. It uses two sensors, one to capture a baseline image with standard exposure, and another to capture multiple images with varying exposures. By analyzing the brightness levels and distribution in the baseline image, the camera determines the optimal exposure time for the second sensor. This allows capturing a blur-reduced image in low light without overexposing or underexposing the entire scene.

Source