This paper investigates the use of a single image of a smooth Lambertian surface to calibrate and remove some image nonlinearities due to the imaging device. To the best of our knowledge, this has not been addressed b...
详细信息
One of the major goals of computervision and machine intelligence is the development of flexible and efficient methods for shape representation. This paper presents an approach for shape retrieval based on sparse rep...
详细信息
This paper presents a novel real-time super-resolution (SR) method using directionally adaptive image interpolation and image restoration. The proposed interpolation method estimates the edge orientation using steerab...
详细信息
ISBN:
(纸本)9781479923427
This paper presents a novel real-time super-resolution (SR) method using directionally adaptive image interpolation and image restoration. The proposed interpolation method estimates the edge orientation using steerable filters and performs edge refinement along the estimated edge orientation. Bi-linear and bi-cubic interpolation filters are then selectively used according to the estimated edge orientation for reducing jagging artifacts in slanting edge regions. The proposed restoration method can effectively remove image degradation caused by interpolation using the directionally adaptive truncated constrained least-squares (TCLS) filter. The proposed method provides high-quality magnified images which are similar to or better than the result of advanced interpolation or SR methods without high computational load. Experimental results indicate that the proposed system gives higher peak-to-peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) values than the state-of-the-art image interpolation methods.
We present an antialiasing method using combined wavelet-Fourier transform and spatially adaptive shrinkage of the transform coefficients. Traditional antialiasing methods employ a simple low-pass filter onto the enti...
详细信息
ISBN:
(纸本)9781479923427
We present an antialiasing method using combined wavelet-Fourier transform and spatially adaptive shrinkage of the transform coefficients. Traditional antialiasing methods employ a simple low-pass filter onto the entire image, so the resulting image loses not only aliasing artifacts but also high-frequency components such as edges and ridges. The proposed algorithm analyzes the property of the LL subband of the discrete wavelet transform (DWT), and reduces aliasing artifacts using patch-adaptive shrinkage of the DWT coefficients. More specifically, an antialiased LL subband is obtained using adaptive patch-based aliasing reduction. To detect an aliased region, we subtract the discrete Fourier transform (DFT) coefficients of the LL subband from the DFT coefficients of antialiased LL subband. The detected aliasing artifacts in the LH, HL, and HH subbands are reduced by patch-wise adaptive shrinkage of the transform coefficients. The resulting antialiased image is obtained using the inverse DWT. The aliasing artifacts can be efficiently reduced by adaptively shrinking wavelet transform coefficients for preserving high-frequency image details. The proposed antialiasing algorithm is suitable for removing aliasing artifacts which frequently occur in imaging sensors with limited resolution.
Abundance fully constrained least squares (FLCS) method has been widely used for spectral unmixing. A modified FCLS (MFCLS) was previously proposed for the same purpose to derive two iterative equations for solving fu...
详细信息
作者：
J. K. ChawM. M. MokjiComputer Vision
Video and Image Processing (CvviP) Laboratory Department of Microelectronic and Computer Engineering Faculty of Electrical Engineering Universiti Teknologi Malaysia Malaysia
Produce recognition system is a system that can categorize types of vegetables and fruits based on features extracted from the images. However, there are numerous features that can be extracted from fruits and vegetab...
Produce recognition system is a system that can categorize types of vegetables and fruits based on features extracted from the images. However, there are numerous features that can be extracted from fruits and vegetables such as colour, texture and shape. As a result, it is effort consuming to identify suitable features ad hoc. Thus, data mining is required to discover the most discriminative features for recognition. This paper aims to extend the usage of data mining algorithm to image domain. Data mining algorithm is preferred to other feature selection algorithms because it discovers nuggets of knowledge that can be understood by human whereas classic feature selection techniques provide outputs that can only be managed by learning algorithms.
作者：
Jinn-Li TanS. A. R Abu-BakarComputer Vision
Video and Image Processing Department of Microelectronics and Computer Engineering Faculty of Electrical Engineering Universiti Teknologi Malaysia Malaysia
This paper presents a license plate character segmentation method in the context of Malaysian cars. First of all, pre-processing steps will enhance the image before Laplacian pyramid takes place. For a proper binariza...
This paper presents a license plate character segmentation method in the context of Malaysian cars. First of all, pre-processing steps will enhance the image before Laplacian pyramid takes place. For a proper binarization, Laplacian pyramid which up-sampled the image from an image lower in the pyramid when the image is captured under low resolution. By using Sobel edge detector and then median filtering, circumscribe rectangle of minimum area is formed and the angle is calculated. At the same time, the area of characters is focussed. The characters are then selected based on connected component analysis after applying Niblack's threshold. Our goal is to segment the characters properly from the steps mentioned. Therefore, our algorithm tries to find the best point to segment the characters using little prior knowledge. Experimental shows promising results on the flexibility of the proposed design method.
Cellular Simultaneous Recurrent Network (CSRN) is a novel bio-inspired recurrent neural network that mimics reinforcement learning in the brain. CSRN has been proven to be a powerful tool for learning and predicting t...
详细信息
ISBN:
(纸本)9781467314886
Cellular Simultaneous Recurrent Network (CSRN) is a novel bio-inspired recurrent neural network that mimics reinforcement learning in the brain. CSRN has been proven to be a powerful tool for learning and predicting temporal information in face image sequences. In this work, we propose a novel implementation of feature-based CSRN for large-scale pose invariant face recognition. We also report systematic evaluation and performance comparison of our feature-based CSRN method with other well-known standard algorithms (PCA, LDA, Bayesian Classifier and EBGM) using face recognition technology standards for large-scale pose invariant face recognition.
Requirement for a person to face a camera for face identification process may no longer be necessary if the face recognition system is robust against variation of facial pose. In this paper, we proposed a face recogni...
详细信息
Requirement for a person to face a camera for face identification process may no longer be necessary if the face recognition system is robust against variation of facial pose. In this paper, we proposed a face recognition method which remains reliable even in very large head pose variations. In this method, feature from local regions of face are extracted after employing both discrete cosine transform and discrete wavelet transform. Learning strategy is then applied to infer the relationship between face in a given pose and its frontal view. Results we obtained are very promising considering that our proposed method solely relies on a single gallery image. We also demonstrated the high performance of our method in a condition whereby the face images are of low-resolution quality.
暂无评论