Feature Detection Scheme Research Articles

A Lightweight Stereo Visual Odometry System for Navigation of Autonomous Vehicles in Low-Light Conditions

Localization of vehicles in a 3D environment is a basic task in autonomous driving. In the low-light environments, it is difficult to navigate independently using a visual odometry for autonomous driving. The main reason for this challenge is the blurred images in the scenes with insufficient illumination. Although numerous works focused on this issue, it still has a number of inherent drawbacks. In this paper, we propose a lightweight stereo visual odometry system for navigation of autonomous vehicles in low-light situations. Contrary to the existing recovery methods, we aim to divide the captured image into the illumination image as well as the reflectance image and only estimate the illumination one, where the enhanced map of the low-light image is acquired by using the retinex theory. In addition, we further utilize a simplified and rapid feature detection scheme, which reduces the computation time by about 85% but maintaining the matching accuracy similar to that of ORB features. Finally, the experiments show that our average memory consumption of our proposed method is much less than the conventional algorithm.

Scale-adaptive local intentional surface feature detection

In this paper, we propose a mesh-based feature detection scheme that focuses on surface features. A class of features of key interest is intentional structures that act as fiducials and that, for instance, can assist in shape retrieval and distortion measurement. We introduce a tunable two-scale depth measurement scheme to quantify the displacement of a vertex from the local surface, which can be a strong indicator of features. We print and scan 3D models with fiducial features appearing across the surface to demonstrate the high fidelity and accuracy of the proposed feature detection scheme. The method outperforms existing 3D feature detection schemes on CAD models and 3D scans alike. We also discuss applications of data embedding enabled by the achievable detection performance.

Cyclostationary Feature Detection Scheme for FBMC and OFDM Cognitive Radio Systems

Cyclostationary Feature Detection Scheme for FBMC and OFDM Cognitive Radio Systems

Registration of fluorescein angiography and optical coherence tomography images of curved retina via scanning laser ophthalmoscopy photographs.

Accurate and automatic registration of multimodal retinal images such as fluorescein angiography (FA) and optical coherence tomography (OCT) enables utilization of supplementary information. FA is a gold standard imaging modality that depicts neurovascular structure of retina and is used for diagnosing neurovascular-related diseases such as diabetic retinopathy (DR). Unlike FA, OCT is non-invasive retinal imaging modality that provides cross-sectional data of retina. Due to differences in contrast, resolution and brightness of multimodal retinal images, the images resulted from vessel extraction of image pairs are not exactly the same. Also, prevalent feature detection, extraction and matching schemes do not result in perfect matches. In addition, the relationships between retinal image pairs are usually modeled by affine transformation, which cannot generate accurate alignments due to the non-planar retina surface. In this paper, a precise registration scheme is proposed to align FA and OCT images via scanning laser ophthalmoscopy (SLO) photographs as intermediate images. For this purpose, first a retinal vessel segmentation is applied to extract main blood vessels from the FA and SLO images. Next, a novel global registration is proposed based on the Gaussian model for curved surface of retina. For doing so, first a global rigid transformation is applied to FA vessel-map image using a new feature-based method to align it with SLO vessel-map photograph, in a way that outlier matched features resulted from not-perfect vessel segmentation are completely eliminated. After that, the transformed image is globally registered again considering Gaussian model for curved surface of retina to improve the precision of the previous step. Eventually a local non-rigid transformation is exploited to register two images perfectly. The experimental results indicate the presented scheme is more precise compared to other registration methods.

Automatic UAV Image Geo-Registration by Matching UAV Images to Georeferenced Image Data

Recent years have witnessed the fast development of UAVs (unmanned aerial vehicles). As an alternative to traditional image acquisition methods, UAVs bridge the gap between terrestrial and airborne photogrammetry and enable flexible acquisition of high resolution images. However, the georeferencing accuracy of UAVs is still limited by the low-performance on-board GNSS and INS. This paper investigates automatic geo-registration of an individual UAV image or UAV image blocks by matching the UAV image(s) with a previously taken georeferenced image, such as an individual aerial or satellite image with a height map attached or an aerial orthophoto with a DSM (digital surface model) attached. As the biggest challenge for matching UAV and aerial images is in the large differences in scale and rotation, we propose a novel feature matching method for nadir or slightly tilted images. The method is comprised of a dense feature detection scheme, a one-to-many matching strategy and a global geometric verification scheme. The proposed method is able to find thousands of valid matches in cases where SIFT and ASIFT fail. Those matches can be used to geo-register the whole UAV image block towards the reference image data. When the reference images offer high georeferencing accuracy, the UAV images can also be geolocalized in a global coordinate system. A series of experiments involving different scenarios was conducted to validate the proposed method. The results demonstrate that our approach achieves not only decimeter-level registration accuracy, but also comparable global accuracy as the reference images.

Feature Detection Scheme Using Cyclic Prefix (CP) in OFDM; Analytical Method for Basic Performance Characteristics and Applications to Mobile Communication Systems

Feature Detection Scheme Using Cyclic Prefix (CP) in OFDM; Analytical Method for Basic Performance Characteristics and Applications to Mobile Communication Systems

Low-Complexity Cyclostationarity Feature Detection Scheme of Localized SC-FDMA Uplink System for Application to Detect and Avoid

This paper proposes a low-complexity cyclostationarity feature detection scheme for detect and avoid (DAA) of Ultra-Wideband (UWB) system in order to solve the coexistence issues between UWB system and Long Term Evolution-Advanced (LTE-Advanced) system. The proposed scheme is suitable for the detection of a localized Single-carrier Frequency Division Multiple Access (SC-FDMA) signal utilized in the uplink of LTE-Advanced system. Compared with conventional cyclostationarity feature detection, the proposed scheme utilizes all possible cyclic-spectrums located in a distributed window function, which is decided by the frequency distribution of the Primary User (PU) signal. The computational complexity of the proposed scheme is low, due to only one window width instead of all occupied spectrum interval will be searched for the possible cyclic-spectrums. On the other hand, the proposed scheme can also avoid the estimation of the cyclic-spectrums when the type of PU signal is unclear or the cyclic-spectrums are hard to estimate. Simulation results indicate that the proposed scheme can make a tradeoff between detection performance and computational complexity. The low-complexity cyclostationarity feature detection also provides a substitute for the energy detection when the later approach suffers from the noise uncertainty and cannot distinguish the target signal type.

Towards an automated ocean feature detection, extraction and classification scheme for SAR imagery

Spaceborne synthetic aperture radar (SAR) observation is an important tool for monitoring and studying changes in various geophysical elements in and above world oceans. Because of SAR's ideal imaging capability and high resolution, the collection of SAR data will likely extend well into the 21st century. As the data become increasingly abundant and computers faster and more affordable, it naturally leads to an increasing need for an automated procedure to replace the labour-intensive manual screening process. In this paper, an integrated scheme for detection, extraction and classification of linear ocean features in SAR imagery is attempted for the purpose of automated screening. The methodology consists of feature detection based on greyscale histogram screening, feature extraction based on two-dimensional wavelet analysis and feature classification based on texture analysis. Using these algorithms on SAR data, several case studies of linear ocean features, including fronts, ice edges and a polar low, are presented herein. Though not fully automated at this stage, the integration of these algorithms seems to lay a promising foundation for the future development of a more automated ocean feature detection, extraction and classification scheme.

New algorithm to phase selection based on wavelet transforms

In long line applications that involve tripping the faulted phase in case of LG faults, single-pole autoreclosure is extensively used. Traditional phase selectors can suffer some deficiencies in their performance due to varying system and fault conditions. However, the advent of wavelet transform techniques with their ability to focus on system transients provides an effective and reliable method for accurate and fast phase selection. This work describes the design and implementation of a new technique using wavelet transform algorithms. The technique is based on using sharp transitions generated on the faulted phase. This paper presents the wavelet transform concept and its value in classification techniques and feature detection schemes. The feature extraction principle is derived. Generated data from the simulation of the 400 kV power system using EMTP were used to test the performance of the technique.

New Algorithm to Phase Selection Based on Wavelet Transforms

In long line applications that involve tripping the faulted phase in case of LG faults, single-pole autoreclosure is extensively used. Traditional phase selectors can suffer some deficiencies in their performance due to varying system and fault conditions. However, the advent of wavelet transform techniques with their ability to focus on system transients provides an effective and reliable method for accurate and fast phase selection. This work describes the design and implementation of a new technique using wavelet transform algorithms. The technique is based on utilizing sharp transitions generated on the faulted phase. The paper presents the wavelet transform concept and its value in classification techniques and feature detection schemes. Feature extraction principle is derived. Generated data from the simulation of 400 kV power system using EMTP were used to test the performance of the technique.

Wavelets and electromagnetic power system transients

The wavelet transform is introduced as a method for analyzing electromagnetic transients associated with power system faults and switching. This method, like the Fourier transform, provides information related to the frequency composition of a waveform, but it is more appropriate than the familiar Fourier methods for the nonperiodic, wide-band signals associated with electromagnetic transients. It appears that the frequency domain data produced by the wavelet transform may be useful for analyzing the sources of transients through manual or automated feature detection schemes. The basic principles of wavelet analysis are set forth, and examples showing the application of the wavelet transform to actual power system transients are presented.

Feature Tracking on the Unit Sphere

Techniques used in a previous study of the objective identification and tracking of meteorological features in model data are extended to the unit sphere. An alternative feature detection scheme is described based on cubic interpolation for the sphere and local maximization. The extension of the tracking technique, used in the previous study, to the unit sphere is described. An example of the application of these techniques to a global relative vorticity field from a model integration are presented and discussed.

Local feature detector using 3-point matching and dynamic programming

A local feature detection scheme for the recognition of occluded objects is developed and reported in this paper. A 3-point matching algorithm is developed to reduce the local feature matching trial. An elastic matching technique using dynamic programming provides the capability of handling the non-linearly distorted features. Also, the proposed scheme is computationally efficient. A set of handtools is selected to evaluate the performance of the algorithm, and the results are encouraging.

Feature-free images

This paper extends earlier work by Morrone and Owens (1987) on the local energy feature detection scheme by explicitly identifying the class of images that are feature-free, that is, for which the local energy function has no local maxima. It is shown that the phase congruency function, which measures the degree of order in the phase spectrum of the image, uniquely represents that image apart from a feature-free image. This results in a demonstration that the features in an image, that is, those points at which the local energy function has a local maximum, are stable with respect to a large class of image variations.