通道修正均衡化的水下图像增强算法

水中介质和微粒的影响导致光波传播衰减和散射, 在成像过程中水下图像会出现模糊和色偏等情况, 这些 水下成像退化的情况给水下的目标识别、目标跟踪、特征提取等应用带来困难. 针对以上问题, 本文提出了一种基 于通道修正均衡化的暗通道先验(CCD)水下图像增强算法: 首先是对色偏的水下图像进行通道修正均衡化, 利用直 方图强度中心做一个映射, 并将映射的三通道信息融合到限制对比度自适应直方图均衡化中, 改善了图像色偏和对 比度不足的情况; 其次是通过暗通道先验算法进行去模糊, 通过水下增强图像数据集的实验表明, CCD比现有算法 更有效地应对了水下图像成像退化问题, 取得了更好的图像质量指标; 此外, 在特征检测预处理步骤中, 本文方法能 够将检测特征点数提高约1.88倍.     

Color-accurate underwater imaging using perceptual adaptive illumination

Capturing color in water is challenging due to the heavy non-uniform attenuation of light in water across the visible spectrum, which results in dramatic hue shifts toward blue. Yet observing color in water is important for monitoring and surveillance as well as marine biology studies related to species identification, individual and group behavior, and ecosystem health and activity monitoring. Underwater robots are equipped with motor control for large scale transects but they lack sensors that enable capturing color-accurate underwater images. We present a method for color-accurate imaging in water called perceptual adaptive illumination. This method dynamically mixes the illumination of an object in a distance-dependent way using a controllable, multi-color light source. The color mix compensates correctly for color loss and results in an image whose color composition is equivalent to rendering the object in air. Experiments were conducted with a color palette in the pool and at three different coral reefs sites, and with an underwater robot collecting image data with the new sensor.     

Color by correlation: a simple, unifying framework for colorconstancy

The paper considers the problem of illuminant estimation: how, given an image of a scene, recorded under an unknown light, we can recover an estimate of that light. Obtaining such an estimate is a central part of solving the color constancy problem. Thus, the work presented will have applications in fields such as color-based object recognition and digital photography. Rather than attempting to recover a single estimate of the illuminant, we instead set out to recover a measure of the likelihood that each of a set of possible illuminants was the scene illuminant. We begin by determining which image colors can occur (and how these colors are distributed) under each of a set of possible lights. We discuss how, for a given camera, we can obtain this knowledge. We then correlate this information with the colors in a particular image to obtain a measure of the likelihood that each of the possible lights was the scene illuminant. Finally, we use this likelihood information to choose a single light as an estimate of the scene illuminant. Computation is expressed and performed in a generic correlation framework which we develop. We propose a new probabilistic instantiation of this correlation framework and show that it delivers very good color constancy on both synthetic and real images. We further show that the proposed framework is rich enough to allow many existing algorithms to be expressed within it: the gray-world and gamut-mapping algorithms are presented in this framework and we also explore the relationship of these algorithms to other probabilistic and neural network approaches to color constancy     

Underwater object detection: architectures and algorithms – a comprehensive review

Underwater object detection is an essential step in image processing and it plays a vital role in several applications such as the repair and maintenance of sub-aquatic structures and marine sciences. Many computer vision-based solutions have been proposed but an optimal solution for underwater object detection and species classification does not exist. This is mainly because of the challenges presented by the underwater environment which mainly include light scattering and light absorption. The advent of deep learning has enabled researchers to solve various problems like protection of the subaquatic ecological environment, emergency rescue, reducing chances of underwater disaster and its prevention, underwater target detection, spooring, and recognition. However, the advantages and shortcomings of these deep learning algorithms are still unclear. Thus, to give a clearer view of the underwater object detection algorithms and their pros and cons, we proffer a state-of-the-art review of different computer vision-based approaches that have been developed as yet. Besides, a comparison of various state-of-the-art schemes is made based on various objective indices and future research directions in the field of underwater object detection have also been proffered.

     

不同颜色模型下自适应直方图拉伸的水下图像增强

目的 水下图像是海洋信息的重要载体,然而与自然环境下的图像相比,其成像原理更复杂、对比度低、可视性差。为保证不同类型水下图像的增强效果,本文提出在两种颜色模型下自适应直方图拉伸的水下图像增强方法。方法 首先,进行基于Gray-World理论对蓝、绿色通道进行颜色均衡化预处理。然后,根据红绿蓝（R-G-B）通道的分布特性和不同颜色光线在水下传播时的选择性衰减,提出基于参数动态优化的R-G-B颜色模型自适应直方图拉伸,并采用引导滤波器降噪。接下来,在CIE-Lab颜色模型,对‘L’亮度和‘a’‘b’色彩分量分别进行线性和曲线自适应直方图拉伸优化。最终,增强的水下图像呈现出高对比度、均衡的饱和度和亮度。结果 选取不同类型的水下图像作为数据集,将本文方法与融合颜色模型（ICM）、非监督颜色纠正模型（UCM）、基于暗通道先验性（DCP）的水下图像复原和基于水下暗通道先验（UDCP）的图像复原方法相比较,增强后的图像具有高对比度和饱和度。定性和定量分析实验结果说明本文提出的方法能够获得更好视觉效果,增强后的图像拥有更高信息熵和较低噪声。结论 在RGB颜色模型中,通过合理地考虑水下图像的分布特性和水下图像退化物理模型提出自适应直方图拉伸方法;在CIE-Lab颜色模型中,引入拉伸函数和指数型曲线函数重分布色彩和亮度两个分量,本方法计算复杂度低,适用于不同复杂环境下的水下图像增强。     

基于颜色衰减先验和白平衡的水下图像复原

由于水体中悬浮颗粒对光线的吸收及散射作用,以及不同波长的光在水下的衰减程度不同,水下图像通常存在细节模糊、对比度低、颜色失真等问题。为改善水下图像质量,提出一种基于颜色衰减先验和白平衡的水下图像复原方法。首先,根据图像在HSV颜色空间中的颜色衰减先验获得场景深度图;其次,利用水下光学衰减特性估计RGB三通道对应的背景光强度与水下透射率,以实现水下图像的清晰化;最后,采用改进的白平衡方法对清晰化后的水下图像进行颜色校正。实验结果表明,该方法可以显著提升水下图像的细节清晰度与颜色保真度,视觉效果更接近自然场景下的图像。     

基于同态滤波的水下图像增强与色彩校正模型

水体对于不同波长的光信号衰减程度不一致,这种现象破坏了水下图像的清晰度和色彩恒定性。为了解决水下图像亮度与色彩扭曲问题,提出一种基于同态滤波的水下图像增强与色彩校正模型。首先,通过比尔-朗伯定律和路径辐射分量构建出水下成像模型。其次,通过同态滤波对未经过衰减的水下图像进行估计。最后,通过麦克劳林级数对水下成像模型进行级数展开,进而推导出一种保持颜色恒定的水下图像色彩校正模型。实验部分分别对比了水下图像的主观视觉效果和客观评价指标,验证了该算法能够有效地保证水下图像的清晰度和色彩恒定性。校正后的水下图像细节丰富,色彩逼真。     

基于衰减补偿与直方图拉伸的水下图像增强算法

水下物理环境复杂多变,导致获取的水下图像颜色失真、对比度低且细节模糊,影响了水下场景探测的准确性。结合衰减补偿和直方图拉伸技术,提出水下图像增强算法ACHS。根据不同颜色通道的衰减特性,设计基于衰减补偿的颜色校正方法解决水下图像颜色失真问题。将需要颜色校正的水下图像从RGB颜色模型转换到LAB颜色模型,使用引导滤波将亮度通道L分解为基础层和细节层,同时提出基于K-means聚类的双直方图增强算法用于增强基础层的对比度,通过Gamma校正突显细节层的纹理结构。在此基础上,累加亮度通道L的基础层和细节层,并将其从LAB颜色模型转换到RGB颜色模型以获取最终的增强图像。实验结果表明,与GDCP、REBE、WaterNet等算法相比,经该算法增强的水下图像可视度较高,并且具有自然的颜色和清晰的细节。     

结构相似性的水下偏振图像复原

目的 针对水下偏振图像存在雾状模糊和场景细节不明显的问题,以水体透射率图与目标反射光图像存在的相互独立性为基础,提出一种基于结构相似性的水下偏振图像复原方法,旨在提高水下偏振图像的清晰度、对比度和色彩真实度。方法 首先,获取同一水下场景下具有正交偏振方向且分别具有最大和最小光强的两幅偏振图像;然后根据透射率图与目标反射光之间的统计无关性,使用结构相似性推导求解透射率的关系式,并通过偏振差分图像计算透射率的初始值,利用该关系式进行水体透射率的迭代求解;最后将透射率代入偏振成像模型得到目标反射光图像,进而进行颜色校正得到复原图像。结果 选取多组正交的水下偏振图像作为研究对象,采用本文提出的方法与另两种偏振复原算法对其进行复原处理,使用对比度、信息熵、灰度平均梯度、峰值信噪比、增强量以及时间等量化指标进行评估。对比实验结果表明,本文算法在对比度、信息熵、灰度平均梯度、增强量以及颜色恢复上都优于另两种偏振图像复原方法,并有较大幅度的提高;灰度平均梯度和对比度较YY算法提高了一倍左右;本文复原图像的色彩分布较均匀使得图像的信息含量大,信息熵高;而突出的EME也证明本文算法的结果纹理清晰、对比度高以及图像复原程度好;提出算法的复原效果有显著的改善,但算法运行时间较长,实时性有待提高。结论 本文基于水下偏振成像模型的分析以及透射率图与目标反射光图像之间的统计无关性,从水体透射率的估计出发进行图像复原,有效地解决了水下偏振图像细节模糊、对比度低的问题。通过对算法实验效果的主客观分析表明,本文算法能有效地复原水下偏振图像,得到对比度高、细节明显和色彩丰富的恢复图像。     

基于暗原色先验原理的水下图像增强

针对水下光衰减和散射导致的图像严重降质问题和用传统方法进行水下图像增强 产生色偏现象,提出一种新的水下图像增强方法。基于暗原色先验原理进行水下图像增强,用 软抠图的方法对图像暗通道进行细化;在图像前0.1%最亮的像素点中,用中值滤波算法计算出 这些像素点的中值,再计算这些像素点和与之对应的中值的差值,差值最小的像素点作为背景 光的预估值,并用该像素点所在区域颜色饱和度方差来判断预估背景光的准确性;利用Retinex 算法和图像各颜色通道的衰减系数比对增强后的图像进行颜色校正。实验表明,该方法能有效 地去除水下图像中的雾色、校正图像色偏问题,进而提高图像对比度。     

水下光学图像重建方法研究进展

水下光学图像可以提供直观丰富的海洋信息,近年来在海洋资源开发、环境保护和海洋工程等诸多领域发挥越来越重要的作用。但是受恶劣复杂的水下成像环境影响,水下光学图像普遍存在对比度低、图像模糊以及颜色失真等质量退化问题,严重制约水下智能处理系统的性能和应用。如何清晰地重建水下光学图像是国内外广泛关注的、具有挑战性的难点问题。随着深度学习技术的蓬勃发展,利用深度学习来提升水下图像质量成为当前的研究热点。鉴于目前国内在水下光学图像重建方面的研究综述较少,本文全面综述其研究进展。分析了水下图像退化机理,总结了现有水下成像模型以及水下图像重建的挑战;梳理了水下光学图像重建方法的发展历程,根据是否采用深度学习以及是否基于成像模型,将现有方法分为4大类,并按照研究发展顺序,依次介绍4类方法的基本思想,分析其优缺点;归纳了目前公开的水下图像数据集以及常用的水下图像质量评价方法,并对8种典型的水下图像重建方法进行了性能评测和对比分析;总结了该领域目前仍存在的问题,展望了后续研究方向,以便于相关研究人员了解该领域的研究现状,促进该领域的技术发展。     

一种欠驱动水下机器人手爪的作业能力研究

在分析当前机器人手爪研究现状的基础上,根据水下作业任务特点,确定了作业型水下机器人多传感器手爪的设计需求,研究了水下手爪原理样机的作业能力,结合典型物体的抓取给出了关键参数关系.该样机可满足在指定环境下进行目标识别、抓取、简单装配、搬运等典型操作任务的要求,为操作型水下机器人多传感器手爪感知系统的研究提供了硬件平台.     

The Retinex based improved underwater image enhancement

The underwater images suffer from low contrast and color distortion due to variable attenuation of light and nonuniform absorption of red, green and blue components. In this paper, we propose a Retinex-based underwater image enhancement approach. First, we perform underwater image enhancement using the contrast limited adaptive histogram equalization (CLAHE), which limits the noise and enhances the contrast of the dark components of the underwater image at the cost of blurring the visual information. Then, in order to restore the distorted colors, we perform the Retinex-based enhancement of the CLAHE processed image. Next, in order to restore the distorted edges and achieve smoothing of the blurred parts of image, we perform bilateral filtering on the Retinex processed image. In order to utilize the individual strengths of CLAHE, Retinex and bilateral filtering algorithms in a single framework, we determine the suitable parameter values. The qualitative and quantitative performance comparison with some of the existing approaches shows that the proposed approach achieves better enhancement of the underwater images.

     

Robust feedback stabilization of underwater robotic vehicles

The controlled motion of an underwater vehicle is very likely to be affected by arbitrary disturbances with considerable magnitudes. In this paper, we develop a simple approach for optimal robust control design of underwater robotic vehicles having decentralized input-output structure. Our design method is based on an explicit condition on the control input matrix which has been found to be necessary and sufficient for a decentralized control system to be robust against arbitrary, but otherwise, bounded disturbances. That makes it possible to get optimal trade-off relations between the bounds of disturbances, the system output accuracy, and the control force limits. For the robust control design purpose, we apply decentralized sliding-mode control the stability of which can be easily verified using Lyapunov theory. In order to show the effectiveness of the design method, the controlled planar motion of an underwater robotic vehicle is taken as an illustrative example.     

Using eigencolor normalization for illumination-invariant color object recognition

Color is one of salient features for color object recognition, however, the colors of object images sensitively depend on scene illumination. To overcome the lighting dependency problem, a color constancy or color normalization method has to be used. This paper presents a color image normalization method, called eigencolor normalization, which consists of two phases as follows. First, the compacting method, which was originally used for compensating the adverse effect due to shape distortion for 2-D planar objects, is exploited for 3-D color space to make the color distribution less correlated and more compact. Second, the compact color image is further normalized by rotating the histogram to align with the reference axis computed. Consequently, the object colors are transformed into a new color space, called eigencolor space, which reflects the inherent colors of the object and is more invariant to illumination changes. Experimental results show that our eigencolor normalization method is superior to other existing color constancy or color normalization schemes on achieving more accurate color object recognition.     

A pair-mode model for underwater single image enhancement

Enhancing underwater images is a challenging problem owing to light scattering and absorption in underwater environments. Such environments provoke several combined degradations in images including color attenuation, blurring and low contrast. Using image processing techniques to enhance this kind of image remains very attractive because of its low-cost of implementation and typically its small number of parameters when compared to more complex learning techniques. This paper proposes an image processing model which first, analyses the color characteristics of the degraded image. Second, decides about the suitable enhancement steps (i.e., mode of operation) to be performed. It operates in two modes (mode-1 and mode-2), both of which investigate a combination of contrast and chromaticity enhancement techniques. The proposed model was tested on 5141 images collected from various, well-known datasets. It was evaluated using eight different measures, some of which are reference-based, and the rest are blind-based. A set of qualitative and quantitative comparisons was conducted, applying more than 20 methods varying between image processing and deep learning. Besides its efficiency and simplicity, the proposed model demonstrates an ability to achieve good contrast ranges, natural-looking colors, and superior or equivalent quality enhancements when compared to other methods.

     

Interactive 3D object recognition pipeline on mobile GPGPU computing platforms using low-cost RGB-D sensors

In this work, we propose the implementation of a 3D object recognition system which will be optimized to operate under demanding time constraints. The system must be robust so that objects can be recognized properly in poor light conditions and cluttered scenes with significant levels of occlusion. An important requirement must be met: The system must exhibit a reasonable performance running on a low power consumption mobile GPU computing platform (NVIDIA Jetson TK1) so that it can be integrated in mobile robotics systems, ambient intelligence or ambient-assisted living applications. The acquisition system is based on the use of color and depth (RGB-D) data streams provided by low-cost 3D sensors like Microsoft Kinect or PrimeSense Carmine. The resulting system is able to recognize objects in a scene in less than 7 seconds, offering an interactive frame rate and thus allowing its deployment on a mobile robotic platform. Because of that, the system has many possible applications, ranging from mobile robot navigation and semantic scene labeling to human–computer interaction systems based on visual information. A video showing the proposed system while performing online object recognition in various scenes is available on our project website (http://www.dtic.ua.es/~agarcia/3dobjrecog-jetsontk1/).     

Color space analysis of mutual illumination

Mutual illumination occurs when light reflected from one surface impinges on a second one. The resulting additional illumination incident on the second surface affects both the color and intensity of the light reflected from it. As a consequence, the image of a surface in the presence of mutual illumination differs from what it otherwise would have been in the absence of mutual illumination. Unaccounted for mutual illumination can easily confuse methods that rely on intensity or color such as shape-from-shading or color-based object recognition. In this correspondence, we introduce an algorithm that removes mutual illumination effects from images. The domain is that of previously-segmented images of convex surfaces of uniform color and diffuse reflectance where for each surface the interreflection occurs mainly from one other surface and can be accurately accounted for within a one-bounce model. The algorithm is based on a singular value decomposition of the colors coming from each surface. Geometrical information about where on the surface the colors emanate from is not required. The RGB triples from a single convex surface experiencing interreflection fall in a plane; intersecting the planes generated from two interreflecting surfaces results in a unique interreflection color. Each pixel can then be factored into its interreflection and no-interreflection components so that a complete no-interreflection image is produced     

Color refinement using deep neural networks for enhancing color recognition in a projector‐camera system

In projector‐camera systems, object recognition is essential to enable users to interact with physical objects. Among several input features used by the object classifier, color information is widely used as it is easily obtainable. However, the color of an object seen by the camera changes due to the projected light from the projector, which degrades the recognition performance. To solve this problem, we propose a method to restore the original color of an object from the observed color through camera. The color refinement method has been developed based on the deep neural network. The inputs to the neural network are the color of the projector light as well as the observed color of the object in multiple color spaces, including RGB, HSV, HIS, and HSL. The neural network is trained in a supervised manner. Through a number of experiments, we show that our refinement method reduces the difference from the original color and improves the object recognition rate implemented with a number of classification methods.