Kaveh Ahmadi

An efficient compression scheme based on adaptive thresholding in wavelet domain using particle swarm optimization

Image compression is one of the most important research areas in the field of image processing due to its large number of applications such as aerial surveillance, reconnaissance, medicine and multimedia communication. Even when high data rates are available, image compression is necessary in order to reduce the transmission cost. For applications involving information security, a fast delivery also reduces the chances of compromise over a communication channel. In this paper, we explore the possibility of using one of the computational intelligence techniques, namely, Particle Swarm Optimization (PSO), for optimal thresholding in the 2-D discrete wavelet transform (DWT) of an image. To this end, a set of optimal thresholds is obtained using the PSO algorithm. Finally, a variable length coding scheme, such as arithmetic coding is used to encode the results. Finding an optimal threshold value for the wavelet coefficients is very crucial in reducing the source entropy and bit-rate reduction. The proposed method is tested using several standard images against other popular techniques and proved to be more efficient compared to other methods. HighlightsAn evolutionary optimization technique is devised for thresholding the wavelet coefficients.PSO is used to find the optimum thresholds for various sub-bands in the wavelet domain.The method is adaptive in the sense that a single threshold is not used for all sub-bands.A VLC coding scheme is used to efficiently compress the wavelet coefficients.

Small dim object tracking using frequency and spatial domain information

Small dim target tracking is an active and important research area in image processing and pattern recognition. Recently, there has been an emphasis on the development of algorithms based on spatial domain Constant False Alarm Rate (CFAR) detection. This paper presents a novel algorithm for detecting and tracking small dim targets in Infrared (IR) image sequences with low Signal to Noise Ratio (SNR) based on the frequency and spatial domain information. Using a Dual-Tree Complex Wavelet Transform (DT-CWT), a CFAR detector is applied in the frequency domain to find potential positions of objects in a frame. Following this step, a Support Vector Machine (SVM) classification is applied to accept or reject each potential point based on the spatial domain information of the frame. The combination of the frequency and spatial domain information demonstrates the high efficiency and accuracy of the proposed method which is supported by the experimental results. A novel algorithm for detecting and tracking small dim targets in Infrared (IR) image sequences is presented.Using a Dual-Tree Complex Wavelet Transform, a CFAR detector is applied to find potential targets.A Support Vector Machine (SVM) classification is applied to refine potential targets.The combination of the frequency and spatial domain information can accurately provide moving object trajectories.

A novel 3D approach for determination of frontal and coronal plane tibial slopes from MR imaging.

BACKGROUND The proximal tibia is geometrically complex, asymmetrical, and variable, is heavily implicated in arthrokinematics of the knee joint, and thus a contributor to knee pathologies such as non-contact anterior cruciate ligament injury. Medial, lateral, and coronal tibial slopes are anatomic parameters that may increase predisposition to knee injuries, but the extent to which each contributes has yet to be fully realized. Previously, two-dimensional methods have quantified tibial slopes, but more reliable 3D methods may prove advantageous. AIMS (1) to explore the reliability of two-dimensional methods, (2) to introduce a novel three-dimensional measurement approach, and (3) to compare data derived from traditional and novel methods. METHODS Medial, lateral, and coronal tibial slope geometry from both knees (left and right) of one subject were obtained via magnetic resonance images and measured by four trained observers from two-dimensional views. The process was repeated via three-dimensional approaches and data evaluated for intra- and inter-rater reliability. RESULTS The conventional method presented a weaker Intraclass Correlation Coefficient (ICC) for the measured slopes (ranging from 0.43 to 0.81) while the resultant ICC for the proposed method indicated greater reliability (ranging from 0.84 to 0.97). Statistical analysis supported the novel approach for production of more reliable and repeatable results for tibial slopes. CONCLUSIONS The novel three-dimensional method for calculating tibial plateau slope may be more reliable than previously established methods and may be applicable in assessment of susceptibility to osteoarthritis, as part of anterior cruciate ligament injury risk assessment, and in total knee implant design.

Small dim object tracking using a multi objective particle swarm optimisation technique

Dim object tracking in a heavy clutter environment is a theoretical and technological challenge in the field of image processing. For a small dim object, conventional tracking methods fail for the lack of geometrical information. Multiple hypotheses testing (MHT) is one of the generally accepted methods in target tracking systems. However, processing a tree structure with a significant number of branches in MHT has been a challenging issue. Tracking high-speed objects with traditional MHT requires some presumptions which limit the capabilities of these methods. This study presents a hierarchal tracking system in two levels to solve this problem. For each point in the lower-level, a multi objective particle swarm optimisation technique is applied to a group of consecutive frames to reduce the number of branches in each tracking tree. Thus, an optimum track for each moving object is obtained in a group of frames. In the upper-level, an iterative process is used to connect the matching optimum tracks of the consecutive frames based on the spatial information and fitness values. The experimental results show that the proposed method has a superior performance in relation to some common dim object tracking methods over different image sequence data sets.

Single-image super resolution using evolutionary sparse coding technique

Sparse coding (SC) has recently become a widely used tool in signal and image processing. The sparse linear combination of elements from an appropriately chosen over-complete dictionary can represent many signal patches. SC applications have been explored in many fields such as image super resolution (SR), image-feature extraction, image reconstruction, and segmentation. In most of these applications, learning-based SC has provided an excellent image quality. SC involves two steps: dictionary construction and searching the dictionary using quadratic programming. This study focuses on the searching step and a new adaptive variation of genetic algorithm is proposed to search and find the optimum closest match in the dictionary. Also, inspired by the proposed evolutionary SC (ESC), a single-image SR algorithm is proposed. A sparse representation for each patch of the low-resolution input image is obtained by ESC and it is used to generate the high-resolution output image. Experimental results show that the proposed ESC-based method would lead to a better SR image quality.

Edge-preserving MRI Super Resolution using a high frequency regularization technique

The resolution of MRI images is limited due to several factors such as imaging hardware or time constraints. However, high MRI image resolution is desired in many medical applications. Traditional Super Resolution (SR) algorithms are generally unable to recover the high frequency (HF) information of MRI images. Recently, spatial adaptive SR algorithms have utilized the combined edge preserving and smoothness constraint methods to improve the quality of the images. Segmenting the image into edge and smooth blocks is a common step which adds to the complexity and execution time of these methods. This paper presents a fast SR technique for MRI images that preserve the edges and improve the visual quality of MRI images without segmenting the images. Experimental results prove the ability of this proposed approach with respect to the traditional SR methods in terms of better visual quality and less execution time.

Social-spider optimised particle filtering for tracking of targets with discontinuous measurement data

The particle filter (PF), a non-parametric implementation of the Bayes filter, is commonly used to estimate the state of a dynamic non-linear non-Gaussian system. The key idea is to construct a posterior probability satisfying a set of hypotheses representing a potential state of the system. Despite PFs successful applications, it suffers from sample impoverishment in real-world applications. Most of the recent PF-based techniques attempt to improve the functionality of the PF through evolutionary algorithms in the cases of unexpected changes in the system states. However, they have not addressed the discontinuity of observations which is unpreventable in the real world. This study incorporates a recently developed social-spider optimisation technique into PF to overcome the drawback of previous methods in these cases. To avoid premature degeneracy, evolutionary search extends the particle search space when observation is unavailable. The social-spider inspired proposal distribution and the corresponding particle weights are derived to approximate real model states. The experimental results show that the proposed method has superior performance in relation to other evolutionary PF in cases of large changes or discontinuous observations.

Automated Measurement of Patient-Specific Tibial Slopes from MRI

Background: Multi-planar proximal tibial slopes may be associated with increased likelihood of osteoarthritis and anterior cruciate ligament injury, due in part to their role in checking the anterior-posterior stability of the knee. Established methods suffer repeatability limitations and lack computational efficiency for intuitive clinical adoption. The aims of this study were to develop a novel automated approach and to compare the repeatability and computational efficiency of the approach against previously established methods. Methods: Tibial slope geometries were obtained via MRI and measured using an automated Matlab-based approach. Data were compared for repeatability and evaluated for computational efficiency. Results: Mean lateral tibial slope (LTS) for females (7.2°) was greater than for males (1.66°). Mean LTS in the lateral concavity zone was greater for females (7.8° for females, 4.2° for males). Mean medial tibial slope (MTS) for females was greater (9.3° vs. 4.6°). Along the medial concavity zone, female subjects demonstrated greater MTS. Conclusion: The automated method was more repeatable and computationally efficient than previously identified methods and may aid in the clinical assessment of knee injury risk, inform surgical planning, and implant design efforts.

A novel Multiple Hypothesis Testing (MHT) scheme for tracking of dim objects

Multiple Hypothesis Tracking (MHT) is an active field for the detection and tracking of low-observable small targets. Most of MHT algorithms are based on a search method in a large tree of possible tracks in a digital image sequence. Though, processing a tree structure with a significant number of branches in MHT has been a challenging issue. Tracking high-speed objects with traditional MHT requires some presumptions which limit the capabilities of these methods. This paper presents a novel MHT system in three steps to solve this problem. The process starts with the root of each track in the video sequence followed by a Particle Swarm Optimization (PSO) search to find the optimum tracks. Iterative process of PSO tries to refine each track. In the third step, the proposed algorithm merges all of the refined tracks related to an object. Efficiency of the proposed method is presented through the figures and tables in the simulation results section.

A new method for the segmentation of algae images using retinex and support vector machine

Bodies of freshwater act as home to many different types of organisms, including algae. These algae can cause harm when something called a harmful algal bloom takes place, and as such it is desired to classify algae in micro-image samples from the freshwater bodies before a bloom occurs. This paper presents a novel method for improving the quality of the algae micro-image and segmenting the algae in the micro-image, two of the steps involved in the automatic recognition and classification of algae in images. First, the algae image quality was improved through the use of the Retinex enhancement technique. Then, the algae in the improved quality image was segmented from the background using a support vector machine. Experimental results indicate that the detection rate of the proposed method is over 95%.