Showing papers on "Real image published in 2014"

From Virtual to Reality: Fast Adaptation of Virtual Object Detectors to Real Domains

Abstract: The most successful 2D object detection methods require a large number of images annotated with object bounding boxes to be collected for training. We present an alternative approach that trains on virtual data rendered from 3D models, avoiding the need for manual labeling. Growing demand for virtual reality applications is quickly bringing about an abundance of available 3D models for a large variety of object categories. While mainstream use of 3D models in vision has focused on predicting the 3D pose of objects, we investigate the use of such freely available 3D models for multicategory 2D object detection. To address the issue of dataset bias that arises from training on virtual data and testing on real images, we propose a simple and fast adaptation approach based on decorrelated features. We also compare two kinds of virtual data, one rendered with real-image textures and one without. Evaluation on a benchmark domain adaptation dataset demonstrates that our method performs comparably to existing methods trained on large-scale real image domains.

Accurate cell segmentation in microscopy images using membrane patterns

Abstract: Motivation: Identifying cells in an image (cell segmentation) is essential for quantitative single-cell biology via optical microscopy. Although a plethora of segmentation methods exists, accurate segmentation is challenging and usually requires problem-specific tailoring of algorithms. In addition, most current segmentation algorithms rely on a few basic approaches that use the gradient field of the image to detect cell boundaries. However, many microscopy protocols can generate images with characteristic intensity profiles at the cell membrane. This has not yet been algorithmically exploited to establish more general segmentation methods. Results: We present an automatic cell segmentation method that decodes the information across the cell membrane and guarantees optimal detection of the cell boundaries on a per-cell basis. Graph cuts account for the information of the cell boundaries through directional cross-correlations, and they automatically incorporate spatial constraints. The method accurately segments images of various cell types grown in dense cultures that are acquired with different microscopy techniques. In quantitative benchmarks and comparisons with established methods on synthetic and real images, we demonstrate significantly improved segmentation performance despite cell-shape irregularity, cell-to-cell variability and image noise. As a proof of concept, we monitor the internalization of green fluorescent proteintagged plasma membrane transporters in single yeast cells. Availability and implementation: Matlab code and examples are available at http://www.csb.ethz.ch/tools/cellSegmPackage.zip. Contact: sotiris.dimopoulos@gmail.com or joerg.stelling@bsse. ethz.ch Supplementary information: Supplementary data are available at Bioinformatics online.

Chaotic image encryption based on circular substitution box and key stream buffer

Abstract: A new image encryption algorithm based on spatiotemporal chaotic system is proposed, in which the circular S-box and the key stream buffer are introduced to increase the security. This algorithm is comprised of a substitution process and a diffusion process. In the substitution process, the S-box is considered as a circular sequence with a head pointer, and each image pixel is replaced with an element of S-box according to both the pixel value and the head pointer, while the head pointer varies with the previous substituted pixel. In the diffusion process, the key stream buffer is used to cache the random numbers generated by the chaotic system, and each image pixel is then enciphered by incorporating the previous cipher pixel and a random number dependently chosen from the key stream buffer. A series of experiments and security analysis results demonstrate that this new encryption algorithm is highly secure and more efficient for most of the real image encryption practices.

A New Iterative Triclass Thresholding Technique in Image Segmentation

Abstract: We present a new method in image segmentation that is based on Otsu's method but iteratively searches for subregions of the image for segmentation, instead of treating the full image as a whole region for processing. The iterative method starts with Otsu's threshold and computes the mean values of the two classes as separated by the threshold. Based on the Otsu's threshold and the two mean values, the method separates the image into three classes instead of two as the standard Otsu's method does. The first two classes are determined as the foreground and background and they will not be processed further. The third class is denoted as a to-be-determined (TBD) region that is processed at next iteration. At the succeeding iteration, Otsu's method is applied on the TBD region to calculate a new threshold and two class means and the TBD region is again separated into three classes, namely, foreground, background, and a new TBD region, which by definition is smaller than the previous TBD regions. Then, the new TBD region is processed in the similar manner. The process stops when the Otsu's thresholds calculated between two iterations is less than a preset threshold. Then, all the intermediate foreground and background regions are, respectively, combined to create the final segmentation result. Tests on synthetic and real images showed that the new iterative method can achieve better performance than the standard Otsu's method in many challenging cases, such as identifying weak objects and revealing fine structures of complex objects while the added computational cost is minimal.

Ghost-Free High Dynamic Range Imaging via Rank Minimization

Abstract: We propose a ghost-free high dynamic range (HDR) image synthesis algorithm using a low-rank matrix completion framework, which we call RM-HDR. Based on the assumption that irradiance maps are linearly related to low dynamic range (LDR) image exposures, we formulate ghost region detection as a rank minimization problem. We incorporate constraints on moving objects, i.e., sparsity, connectivity, and priors on under- and over-exposed regions into the framework. Experiments on real image collections show that the RM-HDR can often provide significant gains in synthesized HDR image quality over state-of-the-art approaches. Additionally, a complexity analysis is performed which reveals computational merits of RM-HDR over recent advances in deghosting for HDR.

FPM: Fine Pose Parts-Based Model with 3D CAD Models

Abstract: We introduce a novel approach to the problem of localizing objects in an image and estimating their fine-pose. Given exact CAD models, and a few real training images with aligned models, we propose to leverage the geometric information from CAD models and appearance information from real images to learn a model that can accurately estimate fine pose in real images. Specifically, we propose FPM, a fine pose parts-based model, that combines geometric information in the form of shared 3D parts in deformable part based models, and appearance information in the form of objectness to achieve both fast and accurate fine pose estimation. Our method significantly outperforms current state-of-the-art algorithms in both accuracy and speed.

Inpainting for Remotely Sensed Images With a Multichannel Nonlocal Total Variation Model

Abstract: Filling dead pixels or removing uninteresting objects is often desired in the applications of remotely sensed images. In this paper, an effective image inpainting technology is presented to solve this task, based on multichannel nonlocal total variation. The proposed approach takes advantage of a nonlocal method, which has a superior performance in dealing with textured images and reconstructing large-scale areas. Furthermore, it makes use of the multichannel data of remotely sensed images to achieve spectral coherence for the reconstruction result. To optimize the proposed variation model, a Bregmanized-operator-splitting algorithm is employed. The proposed inpainting algorithm was tested on simulated and real images. The experimental results verify the efficacy of this algorithm.

Reweighted low-rank matrix recovery and its application in image restoration.

Abstract: In this paper, we propose a reweighted low-rank matrix recovery method and demonstrate its application for robust image restoration. In the literature, principal component pursuit solves low-rank matrix recovery problem via a convex program of mixed nuclear norm and l1 norm. Inspired by reweighted l1 minimization for sparsity enhancement, we propose reweighting singular values to enhance low rank of a matrix. An efficient iterative reweighting scheme is proposed for enhancing low rank and sparsity simultaneously and the performance of low-rank matrix recovery is prompted greatly. We demonstrate the utility of the proposed method both on numerical simulations and real images/videos restoration, including single image restoration, hyperspectral image restoration, and background modeling from corrupted observations. All of these experiments give empirical evidence on significant improvements of the proposed algorithm over previous work on low-rank matrix recovery.

Adaptive color transfer with relaxed optimal transport

Abstract: This paper studies the problem of color transfer between images using optimal transport techniques. While being a generic framework to handle statistics properly, it is also known to be sensitive to noise and outliers, and is not suitable for direct application to images without additional postprocessing regularization to remove artifacts. To tackle these issues, we propose to directly deal with the regularity of the transport map and the spatial consistency of the reconstruction. Our approach is based on the relaxed and regularized discrete optimal transport method of [1]. We extend this work by (i) modeling the spatial distribution of colors within the image domain and (ii) tuning automatically the relaxation parameters. Experiments on real images demonstrate the capacity of our model to adapt itself to the considered data.

Optimal-selection-based suppressed fuzzy c-means clustering algorithm with self-tuning non local spatial information for image segmentation

Abstract: Suppressed fuzzy c-means clustering algorithm (S-FCM) is one of the most effective fuzzy clustering algorithms. Even if S-FCM has some advantages, some problems exist. First, it is unreasonable to compulsively modify the membership degree values for all the data points in each iteration step of S-FCM. Furthermore, duo to only utilizing the spatial information derived from the pixel's neighborhood window to guide the process of image segmentation, S-FCM cannot obtain satisfactory segmentation results on images heavily corrupted by noise. This paper proposes an optimal-selection-based suppressed fuzzy c-means clustering algorithm with self-tuning non local spatial information for image segmentation to solve the above drawbacks of S-FCM. Firstly, an optimal-selection-based suppressed strategy is presented to modify the membership degree values for data points. In detail, during each iteration step, all the data points are ranked based on their biggest membership degree values, and then the membership degree values of the top r ranked data points are modified while the membership degree values of the other data points are not changed. In this paper, the parameter r is determined by the golden section method. Secondly, a novel gray level histogram is constructed by using the self-tuning non local spatial information for each pixel, and then fuzzy c-means clustering algorithm with the optimal-selection-based suppressed strategy is executed on this histogram. The self-tuning non local spatial information of a pixel is derived from the pixels with a similar neighborhood configuration to the given pixel and can preserve more information of the image than the spatial information derived from the pixel's neighborhood window. This method is applied to Berkeley and other real images heavily contaminated by noise. The image segmentation experiments demonstrate the superiority of the proposed method over other fuzzy algorithms.

Image generation apparatus and image generation method

Abstract: A position and rotation information acquisition unit 730 acquires information relating to a position and a rotation of the head of a user who wears a head-mounted display unit 100. A coordinate transformation unit 740 and a panorama image processing unit 750 generate an image to be displayed on the head-mounted display unit using the information relating to the position and the rotation acquired at a certain point of time by the position and rotation information acquisition unit 730. A correction processing unit 780 corrects the generated image using updated information relating to the position and the rotation at a different point of time.

Relative pose estimation for cylinder-shaped spacecrafts using single image

Abstract: This paper describes a novel framework to determine the relative pose and range of a solid-of-revolution-shaped spacecraft from a single image without any artificial beacons. The translation and the symmetry axis of the spacecraft can be estimated from the imaged cross sections of the spacecraft body. Then the pose and range of the spacecraft are fully determined by means of the images of its solar panels and asymmetric feature. Our method has been validated by both synthetic and real images.

License plate recognition

Abstract: A license plate recognition and image review system and processes are described. The system includes grouping of images that are determined to be of the same vehicle, using an image encoded database such that verification of a license plate read is done through comparison of images of the actual vehicle to images from the encoded database and testing of the accuracy of a manual review process by interspersing previously identified images with real images being reviewed in a batch process.

Accurate and robust localization of duplicated region in copy---move image forgery

Abstract: Copy---move image forgery detection has recently become a very active research topic in blind image forensics. In copy---move image forgery, a region from some image location is copied and pasted to a different location of the same image. Typically, post-processing is applied to better hide the forgery. Using keypoint-based features, such as SIFT features, for detecting copy---move image forgeries has produced promising results. The main idea is detecting duplicated regions in an image by exploiting the similarity between keypoint-based features in these regions. In this paper, we have adopted keypoint-based features for copy---move image forgery detection; however, our emphasis is on accurate and robust localization of duplicated regions. In this context, we are interested in estimating the transformation (e.g., affine) between the copied and pasted regions more accurately as well as extracting these regions as robustly by reducing the number of false positives and negatives. To address these issues, we propose using a more powerful set of keypoint-based features, called MIFT, which shares the properties of SIFT features but also are invariant to mirror reflection transformations. Moreover, we propose refining the affine transformation using an iterative scheme which improves the estimation of the affine transformation parameters by incrementally finding additional keypoint matches. To reduce false positives and negatives when extracting the copied and pasted regions, we propose using "dense" MIFT features, instead of standard pixel correlation, along with hysteresis thresholding and morphological operations. The proposed approach has been evaluated and compared with competitive approaches through a comprehensive set of experiments using a large dataset of real images (i.e., CASIA v2.0). Our results indicate that our method can detect duplicated regions in copy---move image forgery with higher accuracy, especially when the size of the duplicated region is small.

ROCHADE: Robust Checkerboard Advanced Detection for Camera Calibration

Abstract: We present a new checkerboard detection algorithm which is able to detect checkerboards at extreme poses, or checkerboards which are highly distorted due to lens distortion even on low-resolution images. On the detected pattern we apply a surface fitting based subpixel refinement specifically tailored for checkerboard X-junctions. Finally, we investigate how the accuracy of a checkerboard detector affects the overall calibration result in multi-camera setups. The proposed method is evaluated on real images captured with different camera models to show its wide applicability. Quantitative comparisons to OpenCV’s checkerboard detector show that the proposed method detects up to 80% more checkerboards and detects corner points more accurately, even under strong perspective distortion as often present in wide baseline stereo setups.

Novel Adaptive Balloon Active Contour Method based on internal force for image segmentation - A systematic evaluation on synthetic and real images

Abstract: Many studies have been conducted on digital image segmentation, seeking to overcome the limitations of different methods for specific applications. Thus, existing techniques are improved and new methods created. This paper proposes a new Active Contour Method (ACM) applied to the segmentation of objects in digital images. The proposed method is called Adaptive Balloon ACM and its main contribution is the new internal Adaptive Balloon energy that minimizes the energy of each point on the curve using the topology of its neighboring points, and thus moves the curve toward the object of interest. The method can be initialized inside or outside the object of interest, and can even segment objects that have complex shapes. There are no limitations as to its startup location. This work evaluates the proposed method in several applications and compares it with other ACMs in the literature. This new method obtained superior results, especially when the objects to be segmented were tubular and had bifurcations. Thus the proposed method can be considered effective in segmenting complex shapes in digital images and gave promising results in various applications.

A novel image segmentation algorithm based on neutrosophic similarity clustering

Abstract: This paper proposed a novel algorithm to segment the objects on images with or without noise.Neutrosophic similarity function is defined to describe the uncertain information on images.A novel objective function is defined using neutrosophic similarity function and the new defined clustering algorithm classifies the pixels on the image into different groups. Segmentation is an important research area in image processing, which has been used to extract objects in images. A variety of algorithms have been proposed in this area. However, these methods perform well on the images without noise, and their results on the noisy images are not good. Neutrosophic set (NS) is a general formal framework to study the neutralities' origin, nature, and scope. It has an inherent ability to handle the indeterminant information. Noise is one kind of indeterminant information on images. Therefore, NS has been successfully applied into image processing algorithms. This paper proposed a novel algorithm based on neutrosophic similarity clustering (NSC) to segment gray level images. We utilize the neutrosophic set in image processing field and define a new similarity function for clustering. At first, an image is represented in the neutrosophic set domain via three membership sets: T, I and F. Then, a neutrosophic similarity function (NSF) is defined and employed in the objective function of the clustering analysis. Finally, the new defined clustering algorithm classifies the pixels on the image into different groups. Experiments have been conducted on a variety of artificial and real images. Several measurements are used to evaluate the proposed method's performance. The experimental results demonstrate that the NSC method segment the images effectively and accurately. It can process both images without noise and noisy images having different levels of noises well. It will be helpful to applications in image processing and computer vision.

MsLRR: A Unified Multiscale Low-Rank Representation for Image Segmentation

Abstract: In this paper, we present an efficient multiscale low-rank representation for image segmentation. Our method begins with partitioning the input images into a set of superpixels, followed by seeking the optimal superpixel-pair affinity matrix, both of which are performed at multiple scales of the input images. Since low-level superpixel features are usually corrupted by image noise, we propose to infer the low-rank refined affinity matrix. The inference is guided by two observations on natural images. First, looking into a single image, local small-size image patterns tend to recur frequently within the same semantic region, but may not appear in semantically different regions. The internal image statistics are referred to as replication prior, and we quantitatively justified it on real image databases. Second, the affinity matrices at different scales should be consistently solved, which leads to the cross-scale consistency constraint. We formulate these two purposes with one unified formulation and develop an efficient optimization procedure. The proposed representation can be used for both unsupervised or supervised image segmentation tasks. Our experiments on public data sets demonstrate the presented method can substantially improve segmentation accuracy.

Real-time causal processing of anomaly detection for hyperspectral imagery

Abstract: Anomaly detection generally requires real-time processing to find targets on a timely basis. However, for an algorithm to be implemented in real time, the used data samples can be only those up to the data sample being visited; no future data samples should be involved in the data processing. Such a property is generally called causality, which has unfortunately received little interest thus far in real-time hyperspectral data processing. This paper develops causal processing to perform anomaly detection that can be also implemented in real time. The ability of real-time causal processing is derived from the concept of innovations used to derive a Kalman filter via a recursive causal update equation. Specifically, two commonly used anomaly detectors, sample covariance matrix (K)-based Reed-Xiaoli detector (RXD), called K-RXD, and sample correlation matrix (R)-based RXD, called R-RXD, are derived for their real-time causal processing versions. To substantiate their utility in applications of anomaly detection, real image data sets are conducted for experiments.

Multiresolution Imaging

Abstract: Imaging resolution has been standing as a core parameter in various applications of vision. Mostly, high resolutions are desirable or essential for many applications, e.g., in most remote sensing systems, and therefore much has been done to achieve a higher resolution of an image based on one or a series of images of relatively lower resolutions. On the other hand, lower resolutions are also preferred in some cases, e.g., for displaying images in a very small screen or interface. Accordingly, algorithms for image upsampling or downsampling have also been proposed. In the above algorithms, the downsampled or upsampled (super-resolution) versions of the original image are often taken as test images to evaluate the performance of the algorithms. However, there is one important question left unanswered: whether the downsampled or upsampled versions of the original image can represent the low-resolution or high-resolution real images from a camera? To tackle this point, the following works are carried out: 1) a multiresolution camera is designed to simultaneously capture images in three different resolutions; 2) at a given resolution (i.e., image size), the relationship between a pair of images is studied, one gained via either downsampling or super-resolution, and the other is directly captured at this given resolution by an imaging device; and 3) the performance of the algorithms of super-resolution and image downsampling is evaluated by using the given image pairs. The key reason why we can effectively tackle the aforementioned issues is that the designed multiresolution imaging camera can provide us with real images of different resolutions, which builds a solid foundation for evaluating various algorithms and analyzing the images with different resolutions, which is very important for vision.

Visual Taxometric Approach to Image Segmentation Using Fuzzy-Spatial Taxon Cut Yields Contextually Relevant Regions

Abstract: Images convey multiple meanings that depend on the context in which the viewer perceptually organizes the scene. By assuming a standardized natural-scene-perception-taxonomy comprised of a hierarchy of nested spatial-taxons [17] [6] [5], image segmentation is operationalized into a series of two-class inferences. Each inference determines the optimal spatial-taxon region, partitioning a scene into a foreground, subject and salient objects and/or sub-objects. I demonstrate the results of a fuzzy-logic-natural-vision-processing engine that implements this novel approach. The engine uses fuzzy-logic inference to simulate low-level visual processes and a few rules of figure-ground perceptual organization. Allowed spatial-taxons must conform to a set of ”meaningfulness” cues, as specified by a generic scene-type. The engine was tested on 70 real images composed of three ”generic scene-types”, each of which required a different combination of the perceptual organization rules built into our model. Five human subjects rated image-segmentation quality on a scale from 1 to 5 (5 being the best). The majority of generic-scene-type image segmentations received a score of 4 or 5 (very good, perfect). ROC plots show that this engine performs better than normalized-cut [9] on generic-scene type images.

Medical Image Segmentation Based on a Hybrid Region-Based Active Contour Model

Abstract: A novel hybrid region-based active contour model is presented to segment medical images with intensity inhomogeneity The energy functional for the proposed model consists of three weighted terms: global term, local term, and regularization term The total energy is incorporated into a level set formulation with a level set regularization term, from which a curve evolution equation is derived for energy minimization Experiments on some synthetic and real images demonstrate that our model is more efficient compared with the localizing region-based active contours (LRBAC) method, proposed by Lankton, and more robust compared with the Chan-Vese (C-V) active contour model

Background subtraction using finite mixtures of asymmetric Gaussian distributions and shadow detection

Abstract: Foreground segmentation of moving regions in image sequences is a fundamental step in many vision systems including automated video surveillance, human-machine interface, and optical motion capture. Many models have been introduced to deal with the problems of modeling the background and detecting the moving objects in the scene. One of the successful solutions to these problems is the use of the well-known adaptive Gaussian mixture model. However, this method suffers from some drawbacks. Modeling the background using the Gaussian mixture implies the assumption that the background and foreground distributions are Gaussians which is not always the case for most environments. In addition, it is unable to distinguish between moving shadows and moving objects. In this paper, we try to overcome these problem using a mixture of asymmetric Gaussians to enhance the robustness and flexibility of mixture modeling, and a shadow detection scheme to remove unwanted shadows from the scene. Furthermore, we apply this method to real image sequences of both indoor and outdoor scenes. The results of comparing our method to different state of the art background subtraction methods show the efficiency of our model for real-time segmentation.

Neural Background Subtraction for Pan-Tilt-Zoom Cameras

Abstract: We propose an extension of a neural-based background subtraction approach to moving object detection to the case of image sequences taken from pan-tilt-zoom (PTZ) cameras. The background model automatically adapts in a self-organizing way to changes in the scene background. Background variations arising in a usual stationary camera setting, such as those due to gradual illumination changes, to waving trees, or to shadows cast by moving objects, are accurately handled by the neural self-organizing background model originally proposed for this type of setting. Handling of variations due to the PTZ camera movement is ensured by a novel registration mechanism that allows the neural background model to automatically compensate the eventual ego-motion, estimated at each time instant. Experimental results on several real image sequences and comparisons with seven state-of-the-art methods demonstrate the accuracy of the proposed approach.

Development and validation of a modelling framework for simulating 2D-mammography and breast tomosynthesis images

Abstract: Planar 2D x-ray mammography is generally accepted as the preferred screening technique used for breast cancer detection. Recently, digital breast tomosynthesis (DBT) has been introduced to overcome some of the inherent limitations of conventional planar imaging, and future technological enhancements are expected to result in the introduction of further innovative modalities. However, it is crucial to understand the impact of any new imaging technology or methodology on cancer detection rates and patient recall. Any such assessment conventionally requires large scale clinical trials demanding significant investment in time and resources. The concept of virtual clinical trials and virtual performance assessment may offer a viable alternative to this approach. However, virtual approaches require a collection of specialized modelling tools which can be used to emulate the image acquisition process and simulate images of a quality indistinguishable from their real clinical counterparts. In this paper, we present two image simulation chains constructed using modelling tools that can be used for the evaluation of 2D-mammography and DBT systems. We validate both approaches by comparing simulated images with real images acquired using the system being simulated. A comparison of the contrast-to-noise ratios and image blurring for real and simulated images of test objects shows good agreement ( < 9% error). This suggests that our simulation approach is a promising alternative to conventional physical performance assessment followed by large scale clinical trials.

Linear Time Distances Between Fuzzy Sets With Applications to Pattern Matching and Classification

Abstract: We present four novel point-to-set distances defined for fuzzy or gray-level image data, two based on integration over α-cuts and two based on the fuzzy distance transform. We explore their theoretical properties. Inserting the proposed point-to-set distances in existing definitions of set-to-set distances, among which are the Hausdorff distance and the sum of minimal distances, we define a number of distances between fuzzy sets. These set distances are directly applicable for comparing gray-level images or fuzzy segmented objects, but also for detecting patterns and matching parts of images. The distance measures integrate shape and intensity/membership of observed entities, providing a highly applicable tool for image processing and analysis. Performance evaluation of derived set distances in real image processing tasks is conducted and presented. It is shown that the considered distances have a number of appealing theoretical properties and exhibit very good performance in template matching and object classification for fuzzy segmented images as well as when applied directly on gray-level intensity images. Examples include recognition of hand written digits and identification of virus particles. The proposed set distances perform excellently on the MNIST digit classification task, achieving the best reported error rate for classification using only rigid body transformations and a kNN classifier.