对于汽车视觉系统的深度学习音色映射和demosaicing
Ana Stojkovic1, Jan Aelterman1, David Van Hamme1
1IMEC, IPI (Image Processing and Interpretation), Ghent University, 9000 Ghent, Belgium.
Sensors (Basel, Switzerland)
|October 28, 2023
概括
两个新的卷积神经网络 (CNN) 将高动态范围 (HDR) 图像转换为8位,改善自动驾驶系统 (ADS) 的对象检测. 这些网络通过在具有挑战性的照明条件下提高检测能力来提高交通参与者的安全.
科学领域:
- 计算机视觉 计算机视觉
- 图像处理 图像处理
- 人工智能的人工智能
背景情况:
- 自动驾驶系统 (ADS) 需要高动态范围 (HDR) 图像,以确保在各种照明条件下安全.
- 目前的对象检测算法是针对8位图像进行了优化,而不是原生HDR更高的比特深度.
- HDR成像和物体检测的独立发展限制了它们的联合有效性.
研究的目的:
- 开发和评估新的卷积神经网络 (CNN) 架构,用于将高位深度HDR图像转换为8位.
- 将HDR优化为8位转换,以提高ADS中的对象检测质量.
- 在重建的8位内容中提高与交通相关的对象的检测能力,同时保持现实主义.
主要方法:
- 提出了两种CNN架构,用于智能HDR到8位图像转换.
- 第一个CNN:在全彩HDR输入上进行联合音色映射和噪声抑制.
- 第二个CNN:在原始HDR输入上进行联合demozaicing,音色映射和噪声抑制.
- 与使用ADS对象检测精度的最先进的音色映射和demosaicing方法进行比较分析.
主要成果:
- 与标准动态范围 (SDR) 内容相比,拟议的CNN在对象检测准确度方面表现出卓越的性能.
- 这些网络在对象检测方面优于现有的最先进的音色映射和解色算法.
- 在重建的8位内容中观察到更好的图像质量和现实主义.
结论:
- 新型CNN有效地将HDR图像转换为8位,以改进ADS对象检测.
- 拟议的方法在自动驾驶中的HDR处理方面比目前的技术有了显著的进步.
- 这项研究通过在具有挑战性的视觉条件下增强感知,为更安全的自动驾驶做出了贡献.
相关概念视频
01:15Depth Perception and Spatial Vision
681
Depth perception is the ability to perceive objects three-dimensionally. It relies on two types of cues: binocular and monocular. Binocular cues depend on the combination of images from both eyes and how the eyes work together. Since the eyes are in slightly different positions, each eye captures a slightly different image. This disparity between images, known as binocular disparity, helps the brain interpret depth. When the brain compares these images, it determines the distance to an object.
681
01:20Deconvolution
168
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...
168
02:16Light Acquisition
8.5K
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.
8.5K
01:24Vision
53.5K
Vision is the result of light being detected and transduced into neural signals by the retina of the eye. This information is then further analyzed and interpreted by the brain. First, light enters the front of the eye and is focused by the cornea and lens onto the retina—a thin sheet of neural tissue lining the back of the eye. Because of refraction through the convex lens of the eye, images are projected onto the retina upside-down and reversed.
53.5K
01:12Differential Leveling
190
Differential leveling is a precise method in surveying used to determine the elevation difference between two points. Its primary goal is to establish accurate vertical measurements to create level surfaces or grade lines critical for designing and constructing infrastructures such as roads, bridges, and buildings.The procedure for differential leveling begins with setting up and leveling the instrument at a point where the benchmark can be seen. The level rod is held on the benchmark (BM), and...
190