The authors propose a line-segment-based marker-less augmented reality (AR) framework that involves an on-site model-generation method and on-line camera tracking. In most conventional model-based marker-less AR frameworks, correspondences between the 3D model and the 2D frame for camera-pose estimation are obtained by feature-point matching. However, 3D models of the target scene are not always available, and feature points are not detected from texture-less objects. The authors’ framework is based on a model-generation method with an RGB-D camera and model-based tracking using line segments, which can be detected even with only a few feature points. The camera pose of the input images can be estimated from the 2D–3D line-segment correspondences given by a line-segment feature descriptor. The experimental results show that the proposed framework can achieve AR when other point-based frameworks cannot. The authors also argue that their framework can generate a model and estimate camera pose more accurately than their previous study.

The authors present a novel method, Hierarchical Manifold Sensing, for adaptive and efficient visual sensing. As opposed to the previously introduced Manifold Sensing algorithm, the new version introduces a way of learning a hierarchical partitioning of the dataset based on k-means clustering. The algorithm can perform on whole images but also on a foveated dataset, where only salient regions are sensed. The authors evaluate the proposed algorithms on the COIL, ALOI, and MNIST datasets. Although they use a very simple nearest-neighbor classifier, on the easier benchmarks, COIL and ALOI, perfect recognition is possible with only six or ten sensing values. Moreover, they show that their sensing scheme yields a better recognition performance than compressive sensing with random projections. On MNIST, state-of-the-art performance cannot be reached, but they show that a large number of test images can be recognized with only very few sensing values. However, for many applications, performance on challenging benchmarks may be less relevant than the simplicity of the solution (processing power, bandwidth) when solving a less challenging problem.

Logarithmic CMOS image sensors are easily able, at video rates, to capture scenes where the dynamic range (DR) is high. However, tone mapping is required to output resulting images or videos to standard low-DR displays. This article proposes a new method, designed especially for logarithmic CMOS image sensors, which can suffer from temporal, and residual fixed pattern, noise. The novel tone mapping, a global operator based on histogram adjustment, uses a model of the camera noise to ensure that the mapping does not amplify the noise above a display threshold. Moreover, to reduce the likelihood of flickering, a temporal adaptation process is incorporated into the histogram calculation. Furthermore, to reduce complexity for real-time processing, a fixed-point implementation is designed for the proposed tone mapping. The novel operator and its fixed-point design are validated through offline and real-time experiments with a logarithmic CMOS image sensor.

With the rapid development of display-related technologies, consumers and manufacturers need an objective measure for numerically rating the relative qualities among different displays. However, existing evaluation measures are unable to satisfy this need, as the results are invariably dissimilar to the results of subjective evaluations as regards display quality. Thus, this study proposes a framework for numerically evaluating the relative qualities among displays based on an analytic network process. To prove the effectiveness of the proposed framework, an experiment is conducted with three different mobile displays, and the results of the proposed framework are compared with those from a subjective evaluation. The relationship between the two sets of data is found to be extremely linear through a correlation coefficient. The authors can confirm that the proposed framework is a good comparative display image quality evaluator.

Feature-level image fusion is investigated with region segmentation and dual-tree complex wavelet transform (DT-CWT). The DT-CWT coefficients are preprocessed by median filtering, and the texture gradients are computed with Gaussian derivatives. The key step of the proposed algorithm is a consistent segmentation of multiple images in the same sub-band. The authors construct a two-order tensor with the gradient of each image, and take the difference between its two eigenvalues as the equivalent gradient squared norm. The watershed algorithm is applied to this norm to obtain the desired region segmentation. Many activity measures of a region are used to construct various fusion rules, and many performance metrics are computed to evaluate the performance. The proposed feature-level approach based on a selection and averaging strategy is compared with many well known approaches, including the latest sparse representation based ones. A comprehensive examination indicates its advantages over other approaches, especially for outdoor applications involving noise and complex background.

This article presents a new spectral reflectance reconstruction method based on Wiener estimation which adaptively reorganizes the training set according to the multi-channel image. First, the initial approximation of the spectral reflectance is reconstructed from the multi-channel images using Wiener estimation. The spectral similarities between each spectral reflectance in the original training set and the initial approximation of the spectral reflectance are then calculated using a correlation coefficient. Next, the similar training set is adaptively organized from the original training set in accordance with each spectral similarity. The final approximation of the spectral reflectance is reconstructed applying the similar training set to Wiener estimation. The performance of the proposed method is objectively compared with previous methods based on the root mean square error and shown to improve the accuracy of the spectral reflectance reconstruction.

City traffic often exhibits regional characteristics, such as large trucks frequently appearing in the suburbs, and the paths to playgrounds on weekends generally being congested. Discovering and visualizing these hidden traffic regions inside which roads share similar characteristics of traffic conditions simplifies the modeling complexities of whole city traffic conditions and therefore contributes significantly toward city planning. Unfortunately, such traffic regions always have irregular shapes and are time varying, which makes their discovery extremely complicated. In addition, establishing a method to visualize and explore the traffic regions interactively still remains challenging. In this article, the authors propose a latent Dirichlet allocation (LDA)-based approach to the discovery of underlying traffic regions (or region topics) from vehicle trajectories captured by surveillance devices installed along roadsides. They treat vehicle trajectories as documents and the values of different traffic features, such as locations, directions, speeds and vehicle types, as the corresponding words. After applying the LDA model, they obtain a list of region topics with combined feature values, in which the different feature values are clustered with probabilistic assignments. Meanwhile, they build a prototype system to explore the surveillance-device-based vehicle trajectories according to the discovered region topics. The prototype system, which consists of map view, cloud view, treemap view and matrix-table view, visualizes the feature values of hidden traffic regions. The authors finally research a real case based on the traffic data in Wenzhou City, a large city in eastern China with a population of more than nine million. They investigate approximately 157 surveillance devices and 750,000 moving vehicles. The case demonstrates the effectiveness of both their proposed approach and the prototype system.