We have witnessed great interest and a wealth of promise in content-based image retrieval as an emerging technology. While the last decade laid foundation to such promise, it also paved the way for a large number of new techniques and systems, got many new people involved, and triggered stronger association of weakly related fields. In this article, we survey almost 300 key theoretical and empirical contributions in the current decade related to image retrieval and automatic image annotation, and in the process discuss the spawning of related subfields. We also discuss significant challenges involved in the adaptation of existing image retrieval techniques to build systems that can be useful in the real world. In retrospect of what has been achieved so far, we also conjecture what the future may hold for image retrieval research.

/pdf/image-retrieval-ideas-influences-and-trends-of-the-new-age-51whw4w7su.pdf

Image retrieval: Ideas, influences, and trends of the new age

State-of-the-art in artificial neural network applications: A survey

A self-scanned 1024 element photodiode array and minicomputer are used to measure the phase (wavefront) in the interference pattern of an interferometer to lambda/100. The photodiode array samples intensities over a 32 x 32 matrix in the interference pattern as the length of the reference arm is varied piezoelectrically. Using these data the minicomputer synchronously detects the phase at each of the 1024 points by a Fourier series method and displays the wavefront in contour and perspective plot on a storage oscilloscope in less than 1 min (Bruning et al. Paper WE16, OSA Annual Meeting, Oct. 1972). The array of intensities is sampled and averaged many times in a random fashion so that the effects of air turbulence, vibrations, and thermal drifts are minimized. Very significant is the fact that wavefront errors in the interferometer are easily determined and may be automatically subtracted from current or subsequent wavefrots. Various programs supporting the measurement system include software for determining the aperture boundary, sum and difference of wavefronts, removal or insertion of tilt and focus errors, and routines for spatial manipulation of wavefronts. FFT programs transform wavefront data into point spread function and modulus and phase of the optical transfer function of lenses. Display programs plot these functions in contour and perspective. The system has been designed to optimize the collection of data to give higher than usual accuracy in measuring the individual elements and final performance of assembled diffraction limited optical systems, and furthermore, the short loop time of a few minutes makes the system an attractive alternative to constraints imposed by test glasses in the optical shop.

Digital wavefront measuring interferometer for testing optical surfaces and lenses

The importance of landscape and heritage recording and documentation with optical remote sensing sensors is well recognized at international level. The continuous development of new sensors, data capture methodologies and multi-resolution 3D representations, contributes significantly to the digital 3D documentation, mapping, conservation and representation of landscapes and heritages and to the growth of research in this field. This article reviews the actual optical 3D measurement sensors and 3D modeling techniques, with their limitations and potentialities, requirements and specifications. Examples of 3D surveying and modeling of heritage sites and objects are also shown throughout the paper.

https://www.mdpi.com/2072-4292/3/6/1104/pdf

Heritage Recording and 3D Modeling with Photogrammetry and 3D Scanning

We present SLIM, a new optical method measuring optical pathlength changes of 0.3 nm spatially and 0.03nm temporally. SLIM combines two classic ideas in light imaging: Zernike’s phase contrast microscopyand Gabor’s holography.

Spatial light interference microscopy (SLIM)

In a method of estimating motion, at least two motion parameter sets are generated (PE1-PEn) from input video data (n, n-1), a motion parameter set being a set of parameters describing motion in an image, by means of which motion parameter set motion vectors can be calculated. One motion parameter set indicates a zero velocity for all image parts in an image, and each motion parameter set has corresponding local match errors. Output motion data are determined from the input video data (n, n-1) in dependence on the at least two motion parameter sets, wherein the importance of each motion parameter set in calculating the output motion data depends on the motion parameter sets' local match errors.

Motion estimation and motion-compensated interpolation

Identifying the materials of a painting is a crucial step in any conservation process. When the objective is to prepare an intervention plan, it is absolutely necessary to understand the matters the restorer is going to encounter. Also, when the aim is a better understanding of the artwork, and perhaps an authenticity check, it is highly relevant to know which materials were employed, since they may differ depending on the period of execution and on the specific artist as well. To identify materials on a painting's surface in a reliable manner, currently the most popular and trustworthy method is the analysis of microsamples of the paint layer. However, chemical analyses although reliable, have a number of drawbacks. The first is linked to the fact that this is an invasive method. The samples used need to be detached from the painting and will not be put back in place. Moreover, the achieved results are in principle - and often also in practice - valid only for that specific specimen and cannot generally be extended to neighboring areas.

Multispectral imaging of paintings

The ability to see behind flames is a key challenge for the industrial field and particularly for the safety field. Development of new technologies to detect live people through smoke and flames in fire scenes is an extremely desirable goal since it can save human lives. The latest technologies, including equipment adopted by fire departments, use infrared bolometers for infrared digital cameras that allow users to see through smoke. However, such detectors are blinded by flame-emitted radiation. Here we show a completely different approach that makes use of lensless digital holography technology in the infrared range for successful imaging through smoke and flames. Notably, we demonstrate that digital holography with a cw laser allows the recording of dynamic human-size targets. In this work, easy detection of live, moving people is achieved through both smoke and flames, thus demonstrating the capability of digital holography at 10.6 μm.

Imaging live humans through smoke and flames using far-infrared digital holography

The application of science and engineering to the analysis of priceless paintings and statues dates back several centuries. However, only over the past few decades have analytical methods developed in the physical sciences been able to glean information, from the past and contribute to the analysis, conservation, and dissemination of works of art. By specifically focusing on the visual arts (e.g., on the binomial image processing and paintings), it is the purpose of this column to: 1) sketch the main applications that could benefit from the introduction of ad hoc image processing tools, 2) outline the state-of-the-art, 3) describe the main peculiarities of this field, and 4) discuss the main obstacles impeding a more fruitful cooperation between the image processing community and art historians, restorers, and artists in general.

Image processing for the analysis and conservation of paintings: opportunities and challenges

In a method of detecting covered and uncovered parts in an image to be interpolated between neighboring previous (I) and next (III) input images, backward motion vectors from the next input image to the previous input image, and having corresponding backward estimation errors, and forward motion vectors from the previous input image to the next input image, and having corresponding forward estimation errors, are determined, uncovered and covered parts are detected in the neighboring previous and next input images, respectively, in the thus detected uncovered parts, uncovered parts in the image to be interpolated are detected by determining second backward estimation errors by comparing both neighboring previous and next input images when partially shifted over the backward motion vectors to a temporal location of said image to be interpolated, and by comparing the second backward estimation errors to a threshold, and in the thus detected covered parts, covered parts in the image to be interpolated are detected by determining second forward estimation errors by comparing both neighboring previous and next input images when partially shifted over the forward motion vectors to a temporal location of the image to be interpolated, and by comparing the second forward estimation errors to the threshold.

Anna Pelagotti

Papers

Motion estimation and motion-compensated interpolation

Multispectral imaging of paintings

Imaging live humans through smoke and flames using far-infrared digital holography

Image processing for the analysis and conservation of paintings: opportunities and challenges

Motion vector estimation and detection of covered/uncovered image parts