Surface based hypothesis verification in intensity images using geometric and appearance data

In this paper we discuss current work concerning Appearance-based and CAD-based vision; two opposing vision strategies. CAD-based vision is geometry based, reliant on having complete object centred models. Appearance-based vision builds view dependent models from training images. Existing CAD-based vision systems that work with intensity images have all used one and zero dimensional features, for example lines, arcs, points and corners. We describe a system we have developed for combining these two strategies. Geometric models are extracted from a commercial CAD library of industry standard parts. Surface appearance characteristics are then learnt automatically by observing actual object instances. This information is combined with geometric information and is used in hypothesis evaluation. This augmented description improves the systems robustness to texture, specularities and other artifacts which are hard to model with geometry alone, whilst maintaining the advantages of a geometric description.

Fusion of infrared polarization and intensity images using support value transform and fuzzy combination rules

Infrared polarization and intensity imagery provide complementary and discriminative information in image understanding and interpretation. In this paper, a novel fusion method is proposed by effectively merging the information with various combination rules. It makes use of both low-frequency and highfrequency images components from support value transform (SVT), and applies fuzzy logic in the combination process. Images (both infrared polarization and intensity images) to be fused are firstly decomposed into low-frequency component images and support value image sequences by the SVT. Then the low-frequency component images are combined using a fuzzy combination rule blending three sub-combination methods of (1) region feature maximum, (2) region feature weighting average, and (3) pixel value maximum; and the support value image sequences are merged using a fuzzy combination rule fusing two sub-combination methods of (1) pixel energy maximum and (2) region feature weighting. With the variables of two newly defined features, i.e. the low-frequency difference feature for low-frequency component images and the support-value difference feature for support value image sequences, trapezoidal membership functions are proposed and developed in tuning the fuzzy fusion process. Finally the fused image is obtained by inverse SVT operations. Experimental results of visual inspection and quantitative evaluation both indicate the superiority of the proposed method to its counterparts in image fusion of infrared polarization and intensity images.

Generating Segmented Quality Meshes from Images

Techniques devoted to generating triangular meshes from intensity images either take as input a segmented image or generate a mesh without distinguishing individual structures contained in the image. These facts may cause difficulties in using such techniques in some applications, such as numerical simulations. In this work we reformulate a previously developed technique for mesh generation from intensity images called Imesh. This reformulation makes Imesh more versatile due to an unified framework that allows an easy change of refinement metric, rendering it effective for constructing meshes for applications with varied requirements, such as numerical simulation and image modeling. Furthermore, a deeper study about the point insertion problem and the development of geometrical criterion for segmentation is also reported in this paper. Meshes with theoretical guarantee of quality can also be obtained for each individual image structure as a post-processing step, a characteristic not usually found in other methods. The tests demonstrate the flexibility and the effectiveness of the approach.

A semi-automatic method for indirect orientation of aerial images using ground control lines extracted from airborne laser scanner data

This paper presents a method for indirect orientation of aerial images using ground control lines extracted from airborne Laser system (ALS) data. This data integration strategy has shown good potential in the automation of photogrammetric tasks, including the indirect orientation of images. The most important characteristic of the proposed approach is that the exterior orientation parameters (EOP) of a single or multiple images can be automatically computed with a space resection procedure from data derived from different sensors. The suggested method works as follows. Firstly, the straight lines are automatically extracted in the digital aerial image (s) and in the intensity image derived from an ALS data-set (S). Then, correspondence between s and S is automatically determined. A line-based coplanarity model that establishes the relationship between straight lines in the object and in the image space is used to estimate the EOP with the iterated extended Kalman filtering (IEKF). Implementation and testing of the method have employed data from different sensors. Experiments were conducted to assess the proposed method and the results obtained showed that the estimation of the EOP is function of ALS positional accuracy.

Towards sea surface pollution detection from visible band images

This paper presents a novel approach to water pollution detection from remotely sensed low-platform mounted visible band camera images. We examine the feasibility of unsupervised segmentation for slick (oily spills on the water surface) region labelling. Adaptive and non adaptive filtering is combined with density modeling of the obtained textural features. A particular effort is concentrated on the textural feature extraction from raw intensity images using filter banks and adaptive feature extraction from the obtained output coefficients. Segmentation in the extracted feature space is achieved using Gaussian mixture models (GMM).

A generative model for separating illumination and reflectance from images

It is well known that even slight changes in nonuniform illumination lead to a large image variability and are crucial for many visual tasks. This paper presents a new ICA related probabilistic model where the number of sources exceeds the number of sensors to perform an image segmentation and illumination removal, simultaneously. We model illumination and reflectance in log space by a generalized autoregressive process and Hidden Gaussian Markov random field, respectively. The model ability to deal with segmentation of illuminated images is compared with a Canny edge detector and homomorphic filtering. We apply the model to two problems: synthetic image segmentation and sea surface pollution detection from intensity images.

The Feasibly of Functionally Guided Intensity Modulated Radiation Therapy Treatment Planning Using Perfluoropropane-Enhanced MRI for Lung Cancer

Radiotherapy is commonly used to treat lung cancer. However, radiation induced damage to lung tissue is a major limiting factor to its use. To minimize normal tissue lung toxicity from conformal radiotherapy treatment planning, we investigated the use of Perfluoropropane(PFP)-enhanced MR imaging to assess and guide the sparing of functioning lung. Fluorine Enhanced MRI using Perfluoropropane(PFP) is a dynamic multi-breath steady state technique enabling quantitative and qualitative assessments of lung function(1).

Imaging data was obtained from studies previously acquired in the Duke Image Analysis Laboratory. All studies were approved by the Duke IRB. The data was de-identified for this project, which was also approved by the Duke IRB. Subjects performed several breath-holds at total lung capacity(TLC) interspersed with multiple tidal breaths(TB) of Perfluoropropane(PFP)/oxygen mixture. Additive wash-in intensity images were created through the summation of the wash-in phase breath-holds. Additionally, model based fitting was utilized to create parametric images of lung function(1).

Varian Eclipse treatment planning software was used for putative treatment planning. For each subject two plans were made, a standard plan, with no regional functional lung information considered other than current standard models. Another was created using functional information to spare functional lung while maintaining dose to the target lesion. Plans were optimized to a prescription dose of 60 Gy to the target over the course of 30 fractions.

A decrease in dose to functioning lung was observed when utilizing this functional information compared to the standard plan for all five subjects. PFP-enhanced MR imaging is a feasible method to assess ventilatory lung function and we have shown how this can be incorporated into treatment planning to potentially decrease the dose to normal tissue.

An integrated approach for precise road reconstruction from aerial imagery and LiDAR data

Accurate and detailed road models play an important role in a number of geospatial applications, such as infrastructure planning, traffic monitoring, and driver assistance systems. In this thesis, an integrated approach for the automatic extraction of precise road features from high resolution aerial images and LiDAR point clouds is presented. A framework of road information modeling has been proposed, for rural and urban scenarios respectively, and an integrated system has been developed to deal with road feature extraction using image and LiDAR analysis. For road extraction in rural regions, a hierarchical image analysis is first performed to maximize the exploitation of road characteristics in different resolutions. The rough locations and directions of roads are provided by the road centerlines detected in low resolution images, both of which can be further employed to facilitate the road information generation in high resolution images. The histogram thresholding method is then chosen to classify road details in high resolution images, where color space transformation is used for data preparation. After the road surface detection, anisotropic Gaussian and Gabor filters are employed to enhance road pavement markings while constraining other ground objects, such as vegetation and houses. Afterwards, pavement markings are obtained from the filtered image using the Otsu's clustering method. The final road model is generated by superimposing the lane markings on the road surfaces, where the digital terrain model (DTM) produced by LiDAR data can also be combined to obtain the 3D road model. As the extraction of roads in urban areas is greatly affected by buildings, shadows, vehicles, and parking lots, we combine high resolution aerial images and dense LiDAR data to fully exploit the precise spectral and horizontal spatial resolution of aerial images and the accurate vertical information provided by airborne LiDAR. Objectoriented image analysis methods are employed to process the feature classiffcation and road detection in aerial images. In this process, we first utilize an adaptive mean shift (MS) segmentation algorithm to segment the original images into meaningful object-oriented clusters. Then the support vector machine (SVM) algorithm is further applied on the MS segmented image to extract road objects. Road surface detected in LiDAR intensity images is taken as a mask to remove the effects of shadows and trees. In addition, normalized DSM (nDSM) obtained from LiDAR is employed to filter out other above-ground objects, such as buildings and vehicles. The proposed road extraction approaches are tested using rural and urban datasets respectively. The rural road extraction method is performed using pan-sharpened aerial images of the Bruce Highway, Gympie, Queensland. The road extraction algorithm for urban regions is tested using the datasets of Bundaberg, which combine aerial imagery and LiDAR data. Quantitative evaluation of the extracted road information for both datasets has been carried out. The experiments and the evaluation results using Gympie datasets show that more than 96% of the road surfaces and over 90% of the lane markings are accurately reconstructed, and the false alarm rates for road surfaces and lane markings are below 3% and 2% respectively. For the urban test sites of Bundaberg, more than 93% of the road surface is correctly reconstructed, and the mis-detection rate is below 10%.

Visual Tracking

A typical robot vision scenario might involve a vehicle moving with an unknown 3D motion (translation and rotation) while taking intensity images of an arbitrary environment. This paper describes the theory and implementation issues of tracking any desired point in the environment. This method is performed completely in software without any need to mechanically move the camera relative to the vehicle. This tracking technique is simple an inexpensive. Furthermore, it does not use either optical flow or feature correspondence. Instead, the spatio-temporal gradients of the input intensity images are used directly. The experimental results presented support the idea of tracking in software. The final result is a sequence of tracked images where the desired point is kept stationary in the images independent of the nature of the relative motion. Finally, the quality of these tracked images are examined using spatio-temporal gradient maps.

Geological image processing of petrographic thin sections using the rotating polarizer stage

One of the fundamental problems with image processing of petrographic thin sections is that the appearance (colour I intensity) of a mineral grain will vary with the orientation of the crystal lattice to the preferred direction of the polarizing filters on a petrographic microscope. This makes it very difficult to determine grain boundaries, grain orientation and mineral species from a single captured image. To overcome this problem, the Rotating Polarizer Stage was used to replace the fixed polarizer and analyzer on a standard petrographic microscope. The Rotating Polarizer Stage rotates the polarizers while the thin section remains stationary, allowing for better data gathering possibilities. Instead of capturing a single image of a thin section, six composite data sets are created by rotating the polarizers through 900 (or 1800 if quartz c-axes measurements need to be taken) in both plane and cross polarized light. The composite data sets can be viewed as separate images and consist of the average intensity image, the maximum intensity image, the minimum intensity image, the maximum position image, the minimum position image and the gradient image. The overall strategy used by the image processing system is to gather the composite data sets, determine the grain boundaries using the gradient image, classify the different mineral species present using the minimum and maximum intensity images and then perform measurements of grain shape and, where possible, partial crystallographic orientation using the maximum intensity and maximum position images.

Stereo vision and wavelets/multiwavelets analysis : 3d reconstruction using wavelets/multiwavelets theory and stereo vision

During the image formation process of the camera, explicit 3D information about the scene or objects in the scene are lost. Therefore, 3D structure or depth information has to be inferred implicitly from the 2D intensity images. This problem is com- monly referred to as 3D reconstruction. In this work a complete 3D reconstruction algorithm is presented, capable of reconstructing dimensionally accurate 3D models of the objects, based on stereo vision and multi-resolution analysis. The developed system uses a reference depth model of the objects under observation to improve the disparity maps, estimated. Only a few features are extracted from that reference model, which are the relative location of the discontinuities and the z-dimensional extremes of objects depth. The maximum error deviation of the estimated depth along the surfaces is less than 0.5mm and along the discontinuities is less than 1.5mm. The developed system is invariant to illuminative variations, and orientation, location and scaling of the objects under consideration, which makes the developed system highly robust.

Hypothesis verification using parametric models and active vision strategies

This paper proposes a methodology for determining the shape and ultimately the functionality of objects from intensity images; 2D analytic functions are used to track 3D features during known camera motions. Three analytic functions are proposed that describe the relationship between pairs of points that are either stationary or moving depending on whether the points are on occluding boundaries or otherwise. Many of the problems of correspondence are reduced by using foveation, known camera motion, and active vision methods. The three analytic functions are shown to enable hypothesis refinement of the functionality of a number of 3D objects without full 3D information about the shape.

A generative probabilistic oriented wavelet model for texture segmentation

This Letter addresses image segmentation via a generative model approach. A Bayesian network (BNT) in the space of dyadic wavelet transform coefficients is introduced to model texture images. The model is similar to a Hidden Markov model (HMM), but with non-stationary transitive conditional probability distributions. It is composed of discrete hidden variables and observable Gaussian outputs for wavelet coefficients. In particular, the Gabor wavelet transform is considered. The introduced model is compared with the simplest joint Gaussian probabilistic model for Gabor wavelet coefficients for several textures from the Brodatz album [1]. The comparison is based on cross-validation and includes probabilistic model ensembles instead of single models. In addition, the robustness of the models to cope with additive Gaussian noise is investigated. We further study the feasibility of the introduced generative model for image segmentation in the novelty detection framework [2]. Two examples are considered: (i) sea surface pollution detection from intensity images and (ii) image segmentation of the still images with varying illumination across the scene.

An integrative approach to face and expression recognition from 3D scans

Biometrics is afield of study which pursues the association of a person's identity with his/her physiological or behavioral characteristics.^ As one aspect of biometrics, face recognition has attracted special attention because it is a natural and noninvasive means to identify individuals. Most of the previous studies in face recognition are based on two-dimensional (2D) intensity images. Face recognition based on 2D intensity images, however, is sensitive to environment illumination and subject orientation changes, affecting the recognition results. With the development of three-dimensional (3D) scanners, 3D face recognition is being explored as an alternative to the traditional 2D methods for face recognition.^ This dissertation proposes a method in which the expression and the identity of a face are determined in an integrated fashion from 3D scans. In this framework, there is a front end expression recognition module which sorts the incoming 3D face according to the expression detected in the 3D scans. Then, scans with neutral expressions are processed by a corresponding 3D neutral face recognition module. Alternatively, if a scan displays a non-neutral expression, e.g., a smiling expression, it will be routed to an appropriate specialized recognition module for smiling face recognition.^ The expression recognition method proposed in this dissertation is innovative in that it uses information from 3D scans to perform the classification task. A smiling face recognition module was developed, based on the statistical modeling of the variance between faces with neutral expression and faces with a smiling expression.^ The proposed expression and face recognition framework was tested with a database containing 120 3D scans from 30 subjects (Half are neutral faces and half are smiling faces). It is shown that the proposed framework achieves a recognition rate 10% higher than attempting the identification with only the neutral face recognition module.^