Fast and Robust Stereo Vision Algorithm for Obstacle Detection

Binocular computer vision is based on bionics, after the calibration through the camera head by double-exposure image synchronization, access to the calculation of two-dimensional image pixels of the three-dimensional depth information. In this paper, a fast and robust stereo vision algorithm is described to perform in-vehicle obstacles detection and characterization. The stereo algorithm which provides a suitable representation of the geometric content of the road scene is described, and an in-vehicle embedded system is presented. We present the way in which the algorithm is used, and then report experiments on real situations which show that our solution is accurate, reliable and efficient. In particular, both processes are fast, generic, robust to noise and bad conditions, and work even with partial occlusion.     

Assessment of vascularity in histological sections: effects of methodology and value as an index of angiogenesis in breast tumours

The aims of this study were to (a) determine how the quantification of blood vessels in histological sections (vascularity) is affected by the methodology used and (b) assess the value of vascularity as an index of angiogenesis by comparing tumour and normal breast tissue. Archival specimens of breast, lung and oral carcinoma, oral dysplasia and normal breast tissue were used to test the effects of the following experimental variables on vascularity: pretreatment of the sections (enzymatic digestion, heating), endothelial markers (von Willebrand factor and CD31 antibodies), method of quantification (highest microvascular density, average microvascular density and microvascular volume) and interobserver variations. All the variables examined significantly affected the estimated vascularity; this depended on the type of tissue and method used. The pretreatment of the sections before staining was the most important variable, altering the vascularity ranking of the tumours. Vascularity in breast tumours was similar to that of the normal breast intralobular stroma, suggesting that an area of high microvascular density in the tumour does not necessarily represent tumour-induced angiogenesis. Contradictory results have been published regarding the value of vascularity as a tumour prognostic factor. Our results suggest that statistically significant differences in vascularity values are most likely to arise from failure to optimize the staining protocol and from the method used to assess vascularity. © 1998 Chapman & Hall     

Thin-section ratiometric Ca2+ images obtained by optical sectioning of fura-2 loaded mast cells

The availability of the ratiometric Ca2+ indicator dyes, fura-2, and indo-1, and advances in digital imaging and computer technology have made it possible to detect Ca2+ changes in single cells with high temporal and spatial resolution. However, the optical properties of the conventional epifluorescence microscope do not produce a perfect image of the specimen. Instead, the observed image is a spatial low pass filtered version of the object and is contaminated with out of focus information. As a result, the image has reduced contrast and an increased depth of field. This problem is especially important for measurements of localized Ca2+ concentrations. One solution to this problem is to use a scanning confocal microscope which only detects in focus information, but this approach has several disadvantages for low light fluorescence measurements in living cells. An alternative approach is to use digital image processing and a deblurring algorithm to remove the out of focus information by using a knowledge of the point spread function of the microscope. All of these algorithms require a stack of two-dimensional images taken at different focal planes, although the "nearest neighbor deblurring" algorithm only requires one image above and below the image plane. We have used a modification of this scheme to construct a simple inverse filter, which extracts optical sections comparable to those of the nearest neighbors scheme, but without the need for adjacent image sections. We have used this "no neighbors" processing scheme to deblur images of fura-2-loaded mast cells from beige mice and generate high resolution ratiometric Ca2+ images of thin sections through the cell. The shallow depth of field of these images is demonstrated by taking pairs of images at different focal planes, 0.5-microns apart. The secretory granules, which exclude the fura-2, appear in focus in all sections and distinct changes in their size and shape can be seen in adjacent sections. In addition, we show, with the aid of model objects, how the combination of inverse filtering and ratiometric imaging corrects for some of the inherent limitations of using an inverse filter and can be used for quantitative measurements of localized Ca2+ gradients. With this technique, we can observe Ca2+ transients in narrow regions of cytosol between the secretory granules and plasma membrane that can be less than 0.5-microns wide. Moreover, these Ca2+ increases can be seen to coincide with the swelling of the secretory granules that follows exocytotic fusion.     

A computational image analysis glossary for biologists

Recent advances in biological imaging have resulted in an explosion in the quality and quantity of images obtained in a digital format. Developmental biologists are increasingly acquiring beautiful and complex images, thus creating vast image datasets. In the past, patterns in image data have been detected by the human eye. Larger datasets, however, necessitate high-throughput objective analysis tools to computationally extract quantitative information from the images. These tools have been developed in collaborations between biologists, computer scientists, mathematicians and physicists. In this Primer we present a glossary of image analysis terms to aid biologists and briefly discuss the importance of robust image analysis in developmental studies.     

Objective evaluation of differences in protein subcellular distribution

Research on the cell and molecular biology of proteins involved in membrane traffic often requires determination of the effects of various experimental conditions on the subcellular distributions of those proteins. This is most often accomplished by acquiring fluorescence microscope images and visually comparing these images. While this approach is quite suitable for detecting major changes in distributions, it is not sensitive to small changes and does not permit a quantitative and objective analysis. We therefore describe the application of pattern analysis methods to the comparison of sets of fluorescence microscope images. This approach provides a high throughput and reproducible technique to determine whether image distributions differ within a specified statistical confidence, and is shown to resolve image sets indistinguishable by visual inspection.     

Image representation using Hermite functions

We present a mathematical technique for analyzing images based on two-dimensional Hermite functions that are translated in both space and spatial frequency. Although the translated functions are not orthogonal, they do constitute a frame and hence can be used for image expansion. The technique has the practical advantage that fast algorithms based on the Zak transform (ZT) can be used to compute expansion coefficients. We describe properties of the ZT that are relevant to image representation and which allow us to use it both to compute expansion coefficients efficiently and to reconstruct images from them. Finally, we use a Hermite function frame to decompose and reconstruct a texture image. Received: 15 June 1993/Accepted in revised form: 22 December 1993     

Multiplexed Nanometric 3D Tracking of Microbeads Using an FFT-Phasor Algorithm

Many single-molecule biophysical techniques rely on nanometric tracking of microbeads to obtain quantitative information about the mechanical properties of biomolecules such as chromatin fibers. Their three-dimensional (3D) position can be resolved by holographic analysis of the diffraction pattern in wide-field imaging. Fitting this diffraction pattern to Lorenz-Mie scattering theory yields the bead’s position with nanometer accuracy in three dimensions but is computationally expensive. Real-time multiplexed bead tracking therefore requires a more efficient tracking method, such as comparison with previously measured diffraction patterns, known as look-up tables. Here, we introduce an alternative 3D phasor algorithm that provides robust bead tracking with nanometric localization accuracy in a z range of over 10 μm under nonoptimal imaging conditions. The algorithm is based on a two-dimensional cross correlation using fast Fourier transforms with computer-generated reference images, yielding a processing rate of up to 10,000 regions of interest per second. We implemented the technique in magnetic tweezers and tracked the 3D position of over 100 beads in real time on a generic CPU. The accuracy of 3D phasor tracking was extensively tested and compared to a look-up table approach using Lorenz-Mie simulations, avoiding experimental uncertainties. Its easy implementation, efficiency, and robustness can improve multiplexed biophysical bead-tracking applications, especially when high throughput is required and image artifacts are difficult to avoid.     

Characterizing Venous Vasculatures of Hepatocellular Carcinoma Using a Multi-Breath-Hold Two-Dimensional Susceptibility Weighted Imaging

The aim of our study is to characterize the venous vasculatures of hepatocellular carcinoma (HCC) using a multi-breath-hold two-dimensional (2D) susceptibility weighted imaging (SWI) in comparison with conventional Magnetic Resonance Imaging (MRI) sequences. Twenty-nine patients with pathologically confirmed HCC underwent MR examination at a 3.0 T scanner. The number of venous vascularity in or around the lesion was counted and the image quality was subjectively evaluated by two experienced radiologists independently based on four image sets: 1) SWI, 2) T1-weighted sequence, 3) T2-weighted sequence, and 4) T1-weighted dynamic contrast-enhanced (DCE) sequence. Of the 29 patients, a total of 33 liver lesions were detected by both SWI and conventional MR sequences. In the evaluation of the conspicuity of venous vascularity, a mean of 10.7 tumor venous vessels per mass was detected by the SWI and 3.9 tumor vasculatures were detected by T1-weighted DCE (P<0.0001), while none was detected by T1-, T2-weighted sequences. The Pearson correlation coefficients between the lesion sizes and the number of tumor vasculatures detected by T1-weighted DCE was 0.708 (P<0.001), and 0.883 by SWI (P<0.001). Our data suggest that SWI appears to be a more sensitive tool compared to T1-weighted DCE sequence to characterize venous vasculature in liver lesions.     

Surface Temperature Distribution of a Breast With and Without Tumour

Breast cancer is a common and dreadful disease in women. Regular screening helps in its early detection. At present the most common methods of screening are by self examination and mammography. The surface temperature distribution of the breast can also provide some information on the presence of tumour. This distribution has a relation to the size and location of tumour and can be seen using thermography, where the infrared radiation emitted from the surface of the breast is recorded and a thermal pattern obtained. Thermography is a non-invasive and an inexpensive tool which could be used for early detection. In order to simulate the surface temperature distribution, a two-dimensional model of female breast with and without a carcinoma is considered. The breast is modelled with varying layer thickness close to the actual shape and numerically solved using finite element analysis. Temperature profiles are obtained for a normal breast and for a malignant one by varying the tumour size, location and the blood flow rates. The results show that the surface temperature for a malignant breast is higher than that of a normal one. In addition the size and location of the tumour do have an effect on the surface temperature distribution. It can also be seen that tumour of different sizes placed at the same location would yield the same maximum temperature depending on the blood perfusion rate.     

Depth Discrimination in Diffuse Optical Transmission Imaging by Planar Scanning Off-Axis Fibers: INITIAL Applications to Optical Mammography

We present a method for depth discrimination in parallel-plate, transmission mode, diffuse optical imaging. The method is based on scanning a set of detector pairs, where the two detectors in each pair are separated by a distance δD_i along direction δ D _i within the x-y scanning plane. A given optical inhomogeneity appears shifted by α_i δ D _i (with 0≤ α_i ≤1) in the images collected with the two detection fibers of the i-th pair. Such a spatial shift can be translated into a measurement of the depth z of the inhomogeneity, and the depth measurements based on each detector pair are combined into a specially designed weighted average. This depth assessment is demonstrated on tissue-like phantoms for simple inhomogeneities such as straight rods in single-rod or multiple-rod configurations, and for more complex curved structures which mimic blood vessels in the female breast. In these phantom tests, the method has recovered the depth of single inhomogeneities in the central position of the phantom to within 4 mm of their actual value, and within 7 mm for more superficial inhomogeneities, where the thickness of the phantom was 65 mm. The application of this method to more complex images, such as optical mammograms, requires a robust approach to identify corresponding structures in the images collected with the two detectors of a given pair. To this aim, we propose an approach based on the inner product of the skeleton images collected with the two detectors of each pair, and we present an application of this approach to optical in vivo images of the female breast. This depth discrimination method can enhance the spatial information content of 2D projection images of the breast by assessing the depth of detected structures, and by allowing for 3D localization of breast tumors.     

Human cortical regions involved in extracting depth from motion

We used functional magnetic resonance imaging (fMRI) to investigate brain regions involved in extracting three-dimensional structure from motion. A factorial design included two-dimensional and three-dimensional structures undergoing rigid and nonrigid motions. As predicted from monkey data, the human homolog of MT/V5 was significantly more active when subjects viewed three-dimensional (as opposed to two-dimensional) displays, irrespective of their rigidity. Human MT/V5+ (hMT/V5+) is part of a network with right hemisphere dominance involved in extracting depth from motion, including a lateral occipital region, five sites along the intraparietal sulcus (IPS), and two ventral occipital regions. Control experiments confirmed that this pattern of activation is most strongly correlated with perceived three-dimensional structure, in as much as it arises from motion and cannot be attributed to numerous two-dimensional image properties or to saliency.     

Serendipia: Castilla-La Mancha telepathology network

Nowadays, there is no standard solution for acquiring, archiving and communication of pathology digital images. In addition, there does not exist any commercial Pathology Information System (LIS) that can manage the relationship between the reports generated by the pathologist and their corresponding images. Due to this situation, the Healthcare Service of Castilla-La Mancha decided to create a completely digital Pathology Department, the project is called SERENDIPIA. SERENDIPIA project provides all the necessary image acquiring devices needed to cover all kind of images that can be generated in a Pathology Department. In addition, in the SERENDIPIA project an Information System was developed that allows it, on the one hand, to cover the daily workflow of a Pathology Department (including the storage and the manage of the reports and its images), and, on the other hand, the Information System provides a WEB telepathology portal with collaborative tools like second opinion.     

An Efficient Object Tracking Method on Quad-/Oc-Trees

We introduce a fast error-free tracking method applicable to sequences of two and three dimensional images. The core idea is to use Quadtree (resp. Octree) data structures for representing the spatial discretization of an image in two (resp. three) spatial dimensions. This representation enables one to merge into large computational cells the regions that can be faithfully described with such a coarse representation, thus significantly reducing the total number of degrees of freedom that are processed, without compromising accuracy. This encoding is particularly effective in the case of algorithms based on moving fronts, since the adaptive refinement provides a natural means to focus the processing resources on information near the moving front. In this paper, we use an existing contour based tracker and reformulate it to the case of Quad-/Oc-tree data structures. Relevant mathematical assumptions and derivations are presented for this purpose. We then demonstrate that, on standard bio-medical image sequences, a speed up of 5X is easily achieved in 2D and about 10X in 3D.     

Surface Temperature Distribution of a Breast With and Without Tumour

Abstract

Breast cancer is a common and dreadful disease in women. Regular screening helps in its early detection. At present the most common methods of screening are by self examination and mammography. The surface temperature distribution of the breast can also provide some information on the presence of tumour. This distribution has a relation to the size and location of tumour and can be seen using thermography, where the infrared radiation emitted from the surface of the breast is recorded and a thermal pattern obtained. Thermography is a non-invasive and an inexpensive tool which could be used for early detection. In order to simulate the surface temperature distribution, a two-dimensional model of female breast with and without a carcinoma is considered. The breast is modelled with varying layer thickness close to the actual shape and numerically solved using finite element analysis. Temperature profiles are obtained for a normal breast and for a malignant one by varying the tumour size, location and the blood flow rates. The results show that the surface temperature for a malignant breast is higher than that of a normal one. In addition the size and location of the tumour do have an effect on the surface temperature distribution. It can also be seen that tumour of different sizes placed at the same location would yield the same maximum temperature depending on the blood perfusion rate.     

Reduction mammaplasty with the "owl" incision and no undermining

Reduction mammaplasty has traditionally been done using the Wise pattern of incision. Because of the box-like effect in breast shape, the lack of projection, and the long scars associated with the inverted T incision, two techniques have emerged as alternatives: the vertical reduction of Lassus/Lejour and the "round block" periareolar technique popularized by Benelli. Each of these techniques has its pros and cons.The "owl" incision combines the features of the large periareolar reduction (Benelli's) and the vertical reduction (Lassus/Lejour); the horizontal inframammary scar is either made very short or completely eliminated. Volume reduction is done through a heart-shaped parenchymal resection, leaving the nipple-areolar complex over a supero-central pedicle. Maintenance of the central parenchyma behind the nipple-areolar complex and mobilization of the vertical pillars toward the center of the breast give excellent projection and diminish the lateral fullness. Enlargement of the periareolar skin resection diminishes the length and pleating of the vertical scar; conversely, inclusion of the vertical component to the periareolar technique eliminates the pleating effect of the periareolar incision. The short horizontal excision eliminates any resultant "dog ears" in the new inframammary fold. Thus, the discrepancy in the length of scars is better distributed. There is no skin or parenchymal undermining, so drains are not needed. Excellent results are obtained immediately on the operating table, and large volumes of glandular resection and correction of severe ptosis can be accomplished without compromising vascularity of either the nipple-areolar complex or the skin flaps.Ninety-four patients in a 7-year period were operated upon using this technique. Seventy-two had bilateral reductions up to 1900 gm per breast, 12 had unilateral reduction for symmetry following breast reconstruction, and 10 were patients with severe ptosis. Complications were rare and of a minor nature. No conversion to free grafts was done, even in the larger resections. One case required minor revision under local anesthesia, one case required bilateral re-reduction, and another case required unilateral re-reduction for continued growth of breast tissue. Almost 90 percent of the patients underwent procedures as outpatients.The owl-type incision and the supero-central pedicle flap are elements of a reduction mammaplasty technique that provides excellent projection and shape with minimal visible scars. It takes advantage of the positive features of the periareolar and vertical reduction techniques and minimizes their negative features. The new design of parenchymal resection improves the vascularity of the residual flaps. Additionally, it may better preserve the sensation to the nipple-areolar complex and lactation is not compromised.     

Automatic Detection of Diseased Tomato Plants Using Thermal and Stereo Visible Light Images

Accurate and timely detection of plant diseases can help mitigate the worldwide losses experienced by the horticulture and agriculture industries each year. Thermal imaging provides a fast and non-destructive way of scanning plants for diseased regions and has been used by various researchers to study the effect of disease on the thermal profile of a plant. However, thermal image of a plant affected by disease has been known to be affected by environmental conditions which include leaf angles and depth of the canopy areas accessible to the thermal imaging camera. In this paper, we combine thermal and visible light image data with depth information and develop a machine learning system to remotely detect plants infected with the tomato powdery mildew fungus Oidium neolycopersici. We extract a novel feature set from the image data using local and global statistics and show that by combining these with the depth information, we can considerably improve the accuracy of detection of the diseased plants. In addition, we show that our novel feature set is capable of identifying plants which were not originally inoculated with the fungus at the start of the experiment but which subsequently developed disease through natural transmission.     

Imaging depth variations in hyperspectral imaging: Development of a method to detect tumor up to the required tumor‐free margin width

Hyperspectral imaging is a promising technique for resection margin assessment during cancer surgery. Thereby, only a specific amount of the tissue below the resection surface, the clinically defined margin width, should be assessed. Since the imaging depth of hyperspectral imaging varies with wavelength and tissue composition, this can have consequences for the clinical use of hyperspectral imaging as margin assessment technique. In this study, a method was developed that allows for hyperspectral analysis of resection margins in breast cancer. This method uses the spectral slope of the diffuse reflectance spectrum at wavelength regions where the imaging depth in tumor and healthy tissue is equal. Thereby, tumor can be discriminated from healthy breast tissue while imaging up to a similar depth as the required tumor‐free margin width of 2 mm. Applying this method to hyperspectral images acquired during surgery would allow for robust margin assessment of resected specimens. In this paper, we focused on breast cancer, but the same approach can be applied to develop a method for other types of cancer.     

Invariant Domain Watermarking Using Heaviside Function of Order Alpha and Fractional Gaussian Field

Digital image watermarking is an important technique for the authentication of multimedia content and copyright protection. Conventional digital image watermarking techniques are often vulnerable to geometric distortions such as Rotation, Scaling, and Translation (RST). These distortions desynchronize the watermark information embedded in an image and thus disable watermark detection. To solve this problem, we propose an RST invariant domain watermarking technique based on fractional calculus. We have constructed a domain using Heaviside function of order alpha (HFOA). The HFOA models the signal as a polynomial for watermark embedding. The watermark is embedded in all the coefficients of the image. We have also constructed a fractional variance formula using fractional Gaussian field. A cross correlation method based on the fractional Gaussian field is used for watermark detection. Furthermore the proposed method enables blind watermark detection where the original image is not required during the watermark detection thereby making it more practical than non-blind watermarking techniques. Experimental results confirmed that the proposed technique has a high level of robustness.