The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In this work we discuss an efficient strategy for reducing the negative impact of non-uniform illumination to panoramic image quality by proposing an adaptive correction algorithm based on the improved Bilateral Gamma function. Firstly the illumination component is extracted by a fast image guided filter. Then an improved bilateral Gamma function fed by the distribution characteristics of illumination...
Segmentation is a fundamental problem in image processing. In biomedical applications, for example cell analysis, it is important to recognize cells with certain shape characteristics. Although recent advances in microscopy and in experimental methods have made the culturing and imaging of cells in 3D environments possible, there is a great need for advanced image processing methods capable of handling...
Remote sensing image registration is still a challenging task because of diverse image types and the lack of a consistent transformation. To improve image registration in remote sensing, this paper develops a robust and accurate feature point matching framework. A modified scale-invariant feature transform (SIFT) method is first introduced for feature detection and pair matching. Based on the properties...
Video Summarization is a computer-based technique to generate a shorter version of the original long video for memory management and information retrieval. The existing method for video summarization applies affective (the state of excitement, interestingness, and panic) level of a viewer captured by Electroencephalography (EEG) while watching a video. The traditional methods for extracting features...
This paper presents a novel local posture orientation-context descriptor, and proposes a FDDL(Fisher discriminant dictionary learning) method based on local orientation-preserving(LOP-FDDL) for sparse coding in action recognition task. To take full use of the information about the position of the local body-part related to the center of the torso, ant the spatial-temporal shape changes of the human...
The traditional procedure for the building change detection is subjective to user's knowledge of the involved data. The complexity increases further due to unavailability of a public data set that is scanned on two different dates. Therefore, the manual changes in the reference data are more common to generate modified data. This paper first presents a strategy to introduce five types of changes in...
Multi-object model-free tracking is challenging because the tracker is not aware of the objects' type (not allowed to use object detectors), and needs to distinguish one object from background as well as other similar objects. Most existing methods keep updating their appearance model individually for each target, and their performance is hampered by sudden appearance change and/or occlusion. We propose...
Classification of Alzheimer 's disease (AD) from normal control (NC) is important for disease abnormality identification and intervention. The current study focused on distinguishing AD from NC based on the multi-feature kernel supervised within- class-similarity discriminative dictionary learning algorithm (MKSCDDL) we introduced previously, which has been derived outperformance in face recognition...
Recently, a lot of works have shown the advantages of utilizing the deep descriptors, obtained from the features of the last convolution layer in CNNs, on image retrieval. In this paper, we focus on augmenting and fusing CNN features for the image retrieval task. We first investigate the effects of network rotation, and then propose two models for deep feature augmenting: single model augmenting and...
This paper presents a new calibration method for lenslet-based plenoptic cameras. While most existing approaches require the computation of sub-aperture images or depth maps which quality depends on some calibration parameters, the proposed process uses the raw image directly. We detect micro-images containing checkerboard corners and use a pattern registration method to estimate their positions with...
Morse code is one of the earliest means of telecommunications; however, it is rarely used nowadays due to viral mobile communications. Although a person can tap Morse codes using his/her fingers easily, perhaps nobody is aware of this kind of finger gestures anymore. In this paper, we will develop a prototype combined together the principle of old Morse code with finger gesture recognition in digital...
Boundary detection in hyperspectral image (HSI) is a challenging task due to high data dimensionality and the that is distributed over the spectral bands. For this reason, there is a dearth of research on boundary detection in HSI. In this paper, we propose a spectral-spatial feature based statistical co-occurrence method for this task. We adopt probability density function (PDF) to estimate the co-occurrence...
In this study, we propose a two-stage method for material segmentation in hyperspectral images. The first stage employs a Convolutional Neural Network (CNN) to predict the material label at individual pixels. The second stage further refines the segmentation by a fully-connected Conditional Random Field (CRF) framework. For the first stage, we experimented with two different network architectures...
Hand-shape recognition is an important problem in computer vision with significant societal impact. In this work, we introduce a new image dataset for Irish Sign Language (ISL) recognition and we compare between two recognition approaches. The dataset was collected by filming human subjects performing ISL hand-shapes and movements. Then, we extracted frames from the videos. This produced a total of...
Color models are widely used in image recognition because they represent significant information. On the other hand, texture analysis techniques have been extensively used for facial feature extraction. In this paper, we extract discriminative features related to facial attributes by utilizing different color models and texture analysis techniques. Specifically, we propose novel methods for texture...
In this study, we introduce an ensemble-based approach for online machine learning. Here, instead of working on the original data, several Hoeffding tree classifiers classify and are updated on the lower dimensional projected data generated from originality by random projections. Since random projection is unstable, from one example, many diverse training data can be created to train the set of Hoeffding...
The maximum consensus problem lies at the core of several important computer vision applications as it is one of the most popular criteria for robust estimation. Although considerable efforts have been devoted to solving this problem, exact algorithms are still impractical for real-world data. Randomized hypothesize-and-test approaches such as RANSAC and its variants are therefore still the key players...
Pedestrian detection for surveillance video, which is the basic of person re-identification, aims to capture the pedestrians in the monitors. However, the existing pedestrian detection algorithms still have two issues: (1) The recall and precision are not applicable for complicated scenes; (2) It is limited for processing the high-resolution video in real-time. Therefore, pedestrian detection algorithm...
We present a method that integrates a part-based sparse appearance model in a Bayesian inference framework for tracking targets in video sequences. We formulate the sparse appearance model as a set of smoothed colour histograms corresponding to the object windows detected by the Deformable Part Model (DPM) detector. The data association of each body part between frames is solved based on the position...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.