使用ColorPix2Pix生成对抗网络在各种天气和照明条件下将图像规范化.
Sanjida Tasnim1, Ashif Mahmud Mostafa1, Azmain Morshed1
1Department of Computer Science and Engineering, BRAC University, Dhaka, Bangladesh.
Scientific reports
|September 30, 2025
概括
本研究介绍了ColorPix2Pix,一种先进的生成对抗网络 (GAN) 算法,用于改善自动驾驶汽车的图像规范化. 新型GAN在不利的照明和天气条件下提高了感知系统的可靠性.
科学领域:
- 计算机视觉 计算机视觉
- 人工智能的人工智能
- 机器人技术 机器人技术 机器人技术
背景情况:
- 自动驾驶汽车 (AV) 依赖于准确的感知系统,以确保安全操作.
- 深度学习对象检测增强了AV感知,但仍然容易受到环境变化的影响,如灯光不佳和恶劣天气.
- 由于环境因素影响图像质量,当前的感知系统在可靠性和安全性方面扎.
研究的目的:
- 开发一种先进的彩色视觉技术,用于正常化在危险的环境和照明条件下捕获的图像.
- 为了引入一种高效的算法,ColorPix2Pix,基于优化的生成对抗网络 (GAN) 模型.
- 在各种环境挑战下提高自动驾驶汽车感知系统的可靠性和安全性.
主要方法:
- 提出了一个新的ColorPix2Pix生成对抗网络 (GAN) 模型,具有增强的损失函数,优先考虑结构和颜色保真性.
- 采用了两阶段的培训过程,使用了模拟雾,雨和可变照明的综合数据集.
- 利用了自定义的损失功能,将感知损失和颜色一致性措施结合起来,以减少噪音和恢复细节.
主要成果:
- ColorPix2Pix GAN有效地使受极端照明和天气条件影响的图像正常化.
- 在照明数据集 (SSIM: 0.767,PSNR: 68.581) 和天气数据集 (SSIM: 0.660,PSNR: 67.185) 上实现了高性能指标.
- 在恢复图像质量和感知可靠性方面,在现有图像规范化方法上表现出卓越的性能.
结论:
- ColorPix2Pix算法显著改善了自动驾驶汽车感知系统的图像正常化.
- 拟议的GAN模型增强了对不利环境条件的强度,这对AV安全至关重要.
- 这项研究通过先进的图像处理技术,有助于实现更可靠,更安全的自动驾驶.
相关概念视频
Color Vision
1.4K
Color perception begins in the retina, the light-sensitive layer at the back of the eye. Two main theories explain how colors are seen: the trichromatic theory and the opponent-process theory. The trichromatic theory, proposed by Thomas Young in 1802 and extended by Hermann von Helmholtz in 1852, suggests that color vision is based on three types of cone receptors in the retina. These cones are sensitive to different but overlapping ranges of wavelengths corresponding to red, blue, and green.
1.4K
Light Acquisition
9.4K
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
9.4K
Photoreceptors and Visual Pathways
8.8K
At the molecular level, visual signals trigger transformations in photopigment molecules, resulting in changes in the photoreceptor cell's membrane potential. The photon's energy level is denoted by its wavelength, with each specific wavelength of visible light associated with a distinct color. The spectral range of visible light, classified as electromagnetic radiation, spans from 380 to 720 nm. Electromagnetic radiation wavelengths exceeding 720 nm fall under the infrared category,...
8.8K
Perceptual Constancy
1.3K
Perceptual constancy is the ability to recognize that objects remain consistent and unchanged even when their appearance varies due to changes in sensory input. There are four main types of perceptual constancy: size constancy, shape constancy, color constancy, and brightness constancy.
Size constancy is the recognition that an object remains the same size, even when its image on the retina changes. For instance, a bus is perceived to be large enough to carry people, even if it looks tiny from...
Size constancy is the recognition that an object remains the same size, even when its image on the retina changes. For instance, a bus is perceived to be large enough to carry people, even if it looks tiny from...
1.3K
Photosystems
6.9K
Photosystems are multiprotein complexes that form the functional units of photosynthesis in plants, algae, and cyanobacteria. They are found embedded in the membrane of tiny sac-like structures called thylakoids placed inside the chloroplast.
Functioning of Photosystems
Photosystems contain many pigment molecules, such as chlorophylls and carotenoids, arranged in a particular organization across two domains — the antenna complex and the reaction center. The main aim of the pigment...
Functioning of Photosystems
Photosystems contain many pigment molecules, such as chlorophylls and carotenoids, arranged in a particular organization across two domains — the antenna complex and the reaction center. The main aim of the pigment...
6.9K
Deconvolution
545
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
545


