Humans perceive the three-dimensional structure of the world with apparent ease. However, despite all of the recent advances in computer vision research, the dream of having a computer interpret an image at the same level as a two-year old remains elusive. Why is computer vision such a challenging problem and what is the current state of the art? Computer Vision: Algorithms and Applications explores the variety of techniques commonly used to analyze and interpret images. It also describes challenging real-world applications where vision is being successfully used, both for specialized applications such as medical imaging, and for fun, consumer-level tasks such as image editing and stitching, which students can apply to their own personal photos and videos. More than just a source of recipes, this exceptionally authoritative and comprehensive textbook/reference also takes a scientific approach to basic vision problems, formulating physical models of the imaging process before inverting them to produce descriptions of a scene. These problems are also analyzed using statistical models and solved using rigorous engineering techniques Topics and features: structured to support active curricula and project-oriented courses, with tips in the Introduction for using the book in a variety of customized courses; presents exercises at the end of each chapter with a heavy emphasis on testing algorithms and containing numerous suggestions for small mid-term projects; provides additional material and more detailed mathematical topics in the Appendices, which cover linear algebra, numerical techniques, and Bayesian estimation theory; suggests additional reading at the end of each chapter, including the latest research in each sub-field, in addition to a full Bibliography at the end of the book; supplies supplementary course material for students at the associated website, http://szeliski.org/Book/. Suitable for an upper-level undergraduate or graduate-level course in computer science or engineering, this textbook focuses on basic techniques that work under real-world conditions and encourages students to push their creative boundaries. Its design and exposition also make it eminently suitable as a unique reference to the fundamental techniques and current research literature in computer vision.

/pdf/computer-vision-algorithms-and-applications-25dn6wu83j.pdf

Computer Vision: Algorithms and Applications

A scheme is developed for classifying the types of motion perceived by a humanlike robot. It is assumed that the robot receives visual images of the scene using a perspective system model. Equations, theorems, concepts, clues, etc., relating the objects, their positions, and their motion to their images on the focal plane are presented. >

http://ieeexplore.ieee.org/iel2/815/3148/00100651.pdf

Robot vision

We present a new technique for the display of high-dynamic-range images, which reduces the contrast while preserving detail. It is based on a two-scale decomposition of the image into a base layer,...

/pdf/fast-bilateral-filtering-for-the-display-of-high-dynamic-43a1vkjsrr.pdf

Fast bilateral filtering for the display of high-dynamic-range images

We present a new technique for the display of high-dynamic-range images, which reduces the contrast while preserving detail. It is based on a two-scale decomposition of the image into a base layer, encoding large-scale variations, and a detail layer. Only the base layer has its contrast reduced, thereby preserving detail. The base layer is obtained using an edge-preserving filter called the bilateral filter. This is a non-linear filter, where the weight of each pixel is computed using a Gaussian in the spatial domain multiplied by an influence function in the intensity domain that decreases the weight of pixels with large intensity differences. We express bilateral filtering in the framework of robust statistics and show how it relates to anisotropic diffusion. We then accelerate bilateral filtering by using a piecewise-linear approximation in the intensity domain and appropriate subsampling. This results in a speed-up of two orders of magnitude. The method is fast and requires no parameter setting.

/pdf/fast-bilateral-filtering-for-the-display-of-high-dynamic-fprymiweg7.pdf

We present a new method for rendering high dynamic range images on conventional displays. Our method is conceptually simple, computationally efficient, robust, and easy to use. We manipulate the gradient field of the luminance image by attenuating the magnitudes of large gradients. A new, low dynamic range image is then obtained by solving a Poisson equation on the modified gradient field. Our results demonstrate that the method is capable of drastic dynamic range compression, while preserving fine details and avoiding common artifacts, such as halos, gradient reversals, or loss of local contrast. The method is also able to significantly enhance ordinary images by bringing out detail in dark regions.

/pdf/gradient-domain-high-dynamic-range-compression-50i9o7k3p7.pdf

Gradient domain high dynamic range compression

A key signal ( alpha ) is generated so that an expression C'=(F-B) alpha ' is satisfied where F, B, and C respectively represent the foreground of a picture which is the part for which the signal ( alpha ) is generated, the background (B) which is the other part than the former part and the value of a picture element constituting the picture and the prime "'" means that the variable marked with "'" is the derivative thereof. A change detecting section (10) calculates the change (C') of the picture element value between picture elements constituting the picture and an F-B detecting section (30) calculates the difference F-B between the foreground (F) and background (B). Then an integrating section (40) finds the key signal ( alpha ) by dividing the change (C') by the difference F-B and integrating the divided results. Therefore, accurate key signals can be obtained by this invention.

Key signal generating device, picture producing device, key signal generating method, and picture producing method

Provided is an imaging device including an image sensor for outputting a high-sensitivity pixel signal from a long-time exposure pixel and a low-sensitivity pixel signal from a short-time exposure pixel; and a blooming correction processing unit for receiving the output pixel signal from the image sensor and executing blooming correction for the pixel signal. The blooming correction processing unit executes pixel value correction of the input pixel from the image sensor according to a difference between or a ratio of an input pixel value from the image sensor and an ideal pixel value not affected by blooming.

Imaging device, image processing method and program for correction of blooming

A smoothing calculator determines a smoothed mean luminance value C1×μ(t)+(1−C1)μ′(t−1) based on a mean luminance value μ(t) input at time t and a smoothed mean luminance value μ′(t−1) read at time (t−1) from a buffer, and outputs the smoothed mean luminance value C1×μ(t)+(1−C1)μ′(t−1) as the smoothed mean luminance value μ(t) of the mean luminance value μ(t). Coefficient C1 is greater than 0 but smaller than 1. Gradation conversion responsive to luminance variation from frame to frame in a moving image is thus achieved.

Method apparatus, and recording medium for smoothing luminance of an image

The present invention relates to an image processing apparatus which can restore, from a color and sensitivity mosaic image acquired using a CCD image sensor of the single plate type or the like, a color image signal of a wide dynamic range wherein the sensitivity characteristics of pixels are uniformized and each of the pixels has all of a plurality of color components. A sensitivity uniformization section uniformizes the sensitivities of pixels of a color and sensitivity mosaic image to produce a color mosaic image, and a color interpolation section interpolates color components of the pixels of the color mosaic image M to produce output images R, G and B. The present invention can be applied to a digital camera which converts a picked up optical image into a color image signal of a wide dynamic range.

Image processing apparatus and method for generating a restoration image

PROBLEM TO BE SOLVED: To obtain a correct key signal. SOLUTION: In an image, the section defined as an object to generate a key signal α is defined as a foreground F, the other section is defined as a background B and the picture element value of picture elements consisting of that image is defined as C. At such a time, the key signal is generated so as to satisfy an expression C'=(F-B)α' (''' expresses differentiation in this case). Namely, at a change detection part 1, a change C' in the picture element values among the picture elements consisting of the image is detected and at an F-B detection part 3, difference F-B between the foreground and the background is calculated. At an integrated value calculation part 4, C' is divided with F-B and that divided result is integrated so that the key signal α can be obtained.

Tomoo Mitsunaga

Papers

Key signal generating device, picture producing device, key signal generating method, and picture producing method

Imaging device, image processing method and program for correction of blooming

Method apparatus, and recording medium for smoothing luminance of an image

Image processing apparatus and method for generating a restoration image

Key signal generator, image composing device, key signal generating method and image composing method