Usman Ullah Sheikh

Near infrared face recognition using Zernike moments and Hermite kernels

This work proposes a novel face recognition method based on Zernike moments (ZMs) and Hermite kernels (HKs) to cope with variations in facial expression, changes in head pose and scale, occlusions due to wearing eyeglasses and the effects of time lapse. Near infrared images are used to tackle the impact of illumination changes on face recognition, and a combination of global and local features is utilized in the decision fusion step. In the global part, ZMs are used as a feature extractor and in the local part, the images are partitioned into multiple patches and filtered patch-wise with HKs. Finally, principal component analysis followed by linear discriminant analysis is applied to data vectors to generate salient features and decision fusion is applied on the feature vectors to properly combine both global and local features. Experimental results on CASIA NIR and PolyU NIR face databases clearly show that the proposed method achieves significantly higher face recognition accuracy compared with existing methods.

3D shape descriptor for object recognition based on Kinect-like depth image

3D shape descriptor has been used widely in the field of 3D object retrieval. However, the performance of object retrieval greatly depends on the shape descriptor used. The aims of this study is to review and compare the common 3D shape descriptors proposed in 3D object retrieval literature for object recognition and classification based on Kinect-like depth image obtained from RGB-D object dataset. In this paper, we introduce (1) inter-class; and (2) intra-class evaluation in order to study the feasibility of such descriptors in object recognition. Based on these evaluations, local spin image outperforms the rest in discriminating different classes when several depth images from an instance per class are used in inter-class evaluation. This might be due to the slightly consistent local shape property of such images and due to the proposed local similarity measurement that manages to extract the local based descriptor. However, shape distribution performs excellent for intra-class evaluation (that involves several instances per class) may be due to the global shape from different instances per class is slightly unchanged. These results indicate a remarkable feasibility analysis of the 3D shape descriptor in object recognition that can be potentially used for Kinect-like sensor.

Rotation and noise invariant near-infrared face recognition by means of Zernike moments and spectral regression discriminant analysis

Abstract. Face recognition is a rapidly growing research area, which is based heavily on the methods of machine learning, computer vision, and image processing. We propose a rotation and noise invariant near-infrared face-recognition system using an orthogonal invariant moment, namely, Zernike moments (ZMs) as a feature extractor in the near-infrared domain and spectral regression discriminant analysis (SRDA) as an efficient algorithm to decrease the computational complexity of the system, enhance the discrimination power of features, and solve the “small sample size” problem simultaneously. Experimental results based on the CASIA NIR database show the noise robustness and rotation invariance of the proposed approach. Further analysis shows that SRDA as a sophisticated technique, improves the accuracy and time complexity of the system compared with other data reduction methods such as linear discriminant analysis.

Moving object detection using image registration for a moving camera platform

This paper proposes an accurate moving object detector from Unmanned Aerial Platform (UAV). Due to the distance of UAV to targets and the movement of the platform, object detection is a challenging task. In order to achieve the best result with low error, at first the camera motion has to be estimated; therefore, by using the Shi & Tomasi corner detector the corners are detected. Using image registration the motion is estimated and then compensated. After motion compensation, adaptive background subtraction is applied for detecting and extracting the moving objects. Experiments on different data sets show the accuracy of our approach on detecting moving objects from aerial platform with up to 85 % correct detection.

Vehicle speed detection using frame differencing for smart surveillance system

This paper presents vehicle speed detection algorithm and its application for smart surveillance system using PC workstation at the selected road lane. Typical temporal resolution of 30 ms to 40ms of conventional camera has been utilized. The resolution is limited to the frame-rate of camera where the maximum velocity that can be computed depends on two factors: the timing resolution and the displacement resolution in pixels on image. The speed estimation of car and motorcycle are obtained in real-time using frame differencing technique for moving object detection. Experimental results are presented to demonstrate the efficiency of the proposed algorithm where it has been tested on a real video sequence and the results are analyzed.

Near Infrared Face Recognition: A Comparison of Moment-Based Approaches

Moment based methods have evolved into a powerful tool for face recognition applications. In this paper, a comparative study on moments based feature extraction methods in terms of their capability to recognize facial images with different challenges is done to evaluate the performance of different type of moments. The moments include Geometric moments (GM’s), Zernike moments (ZM’s), Pseudo-Zernike moments (PZM’s) and Wavelet moments (WM’s). Experiments conducted on CASIA NIR database showed that Zernike moments outperformed other moment-based methods for facial images with different challenges such as facial expressions, head pose and noise.

An adapted point based tracking for vehicle speed estimation in linear spacing

Vehicle velocity estimation is an important aspect of intelligent transportation systems. Normally velocity is estimated using dedicated laser speed traps and Doppler radars. Recently, the use of cameras is becoming more common for the purpose of traffic surveillance and smart surveillance system. It is thus the aim of this paper to propose a method for vehicle speed estimation using these existing video cameras. In this paper, we propose a vehicle speed estimation method from video analysis. The method proposed contains several steps; image preprocessing, centroid extraction and tracking. The proposed method transforms the 2D image points into a 3D virtual world to obtain actual vehicle position in 3D space. This is to account for perspective distortion commonly seen in images. Using these 3D points and measuring the time for displacement, the vehicle speed is obtained. Experimental results have shown that the proposed method gives accurate velocity estimation.

2-D DWT System Architecture for Image Compression

Wavelet transform has contributed significantly in multiple areas such as image processing, compression, signal analysis, and medical imaging. Discrete wavelet transform (DWT) requires very large memory requirement and is computationally intensive, especially for 2-D transform that has a quadratic computational complexity. In this paper, we propose a dedicated processor for 2-D DWT computation. The DWT system architecture is parameterizable, where its performance can be scaled by increasing or reducing the DWT engines, according to different application needs. This architecture requires significantly less computational resources and internal memory. The proposed architecture can achieve a theoretical throughput of 138 frames per second for a 2048 × 1536 video processing. The DWT system has been designed for scalability to support up to 8 parallel DWT engines.

Restoration of out of Focus Barcode Images Using Wiener Filter

Barcode capturing using embedded camera in mobile phone is prone to producing images with out of focus blurring. Barcode decoding systems usually fail to decode barcodes from an image that is affected by image blurring. With macro function, users are able to capture sharp barcodes. Unfortunately, low end and mid range cameras usually do not have macro function which enables close range image capturing. To improve barcode recognition, we propose a technique to restore barcode images before the decoding process. First, we identify and calculate the point spread function (PSF) parameter and radius of the blur in Cepstrum domain. Then we restore the degraded image using Wiener filter. Experiments on both synthetic and real images show that our technique is adept in estimating the radius of the blur. The Wiener filter gives satisfactory result when it is used to restore blurred images with the acquired PSF from the preceding step.

Near infrared face recognition

As a primary modality in biometrics, human face recognition has been employed widely in the computer vision domain because of its performance in a wide range of applications such as surveillance systems and forensics. Recently, near infrared (NIR) imagery has been used in many face recognition systems because of the high robustness to illumination changes in the acquired images. Even though some surveys have been conducted in this infrared domain, they have focused on thermal infrared methods rather than NIR methods. Furthermore, none of the previous infrared surveys provided comprehensive and critical analyses of NIR methods. Therefore, this paper presents an up-to-date survey of the well-known NIR methods that are used to solve the problem of illumination. The paper includes a discussion of the benefits and drawbacks of various NIR methods. Finally, the most promising avenues for future research are highlighted.