Sharif Digital Repository

Color image segmentation is a primitive operation in many image processing and computer vision applications. Accordingly, there exist numerous segmentation approaches in the literature, which might be misleading for a researcher who is looking for a practical algorithm. While many researchers are still using the tools which belong to the old color space paradigm, there is evidence in the research established in the eighties that a proper descriptor of color vectors should act locally in the color domain. In this paper, these results are used to propose a new color image segmentation method. The proposed method searches for the principal colors, defined as the intersections of the cylindrical... 

A new single-pixel Fourier photoacoustic microscopy (PAM), to the best of our knowledge, is proposed to improve the resolution and region of interest (ROI) of an acquired image. In the previous structure of single-pixel Fourier PAM, called spatially invariant resolution PAM (SIR-PAM), the lateral resolution and ROI are limited by the digital micromirror device (DMD) pixel size and the number of pixels. This limitation is overcome here through illuminating fixed angle interfering plane waves, changing the fringe frequency via varying the frequency of the laser source. Given that the fringe sinusoidal patterns here can be produced by two mirrors, the DMD usage can be omitted. In this way, the... 

Defining a suitable similarity measure is a crucial step in (medical) image registration tasks. A common problem with frequently used intensity-based image registration algorithms is that they assume intensities of different pixels are independent of each other that could lead to low registration performance especially in the presence of spatially-varying intensity distortions, because they ignore the complex interactions between the pixel intensities. Motivated by this problem, in this paper we present a novel similarity measure which takes into account nonstationarity of the pixels intensity and complex spatially varying intensity distortions in mono-modal settings. Experimental results on... 

Analysis of microparticle size and fluorescence intensity can be used to classify microparticles. We designed and fabricated an optofluidic system that characterizes microparticles, including fluorescent microparticles and microalgae. A new type of multi-pixel photon counter (MPPC) was employed to miniaturize the device, lower its power consumption, and make it insensitive to magnetic fields. The system uses a 635 nm laser for excitation of the microparticles' fluorescence. The scattered light from the fluorescent microparticles, as well as Chlorella vulgaris and Chlorella protothecoides, were measured. Additionally, we analyzed the width and height of the measured signals generated as a... 

The variation of pose, illumination, and expression continues to make face recognition a challenging problem. As a pre-processing step in holistic approaches, faces are usually aligned by eyes. The proposed method tries to perform a pixel alignment rather than eye alignment by mapping the geometry of faces to a reference face while keeping their own textures. The proposed geometry alignment not only creates a meaningful correspondence among every pixel of all faces, but also removes expression and pose variations effectively. The geometry alignment is performed pixel-wise, i.e., every pixel of the face is corresponded to a pixel of the reference face. In the proposed method, the information... 

In this paper, a design and construction method for an omnidirectional vision system is described, including how to use it on autonomous soccer robots for object detection, localization and, also, collision avoidance in the middle size league of RoboCup. This vision system uses two mirrors, flat and hyperbolic. The flat mirror is used for detecting very close objects around the robot body and the hyperbolic one is used as a global viewing device to construct a world model for the soccer field. This world model contains information about the position and orientation of the robot itself and the position of other objects in a fixed coordinate system. In addition, a fast object detection method... 

Determining the capacity of image for data hiding is a complex task. This capacity may be influenced by many factors such as the content of image and also Human Visual System (HVS). A few works on this topic are reported. In this paper we propose a new method for calculating capacity in gray scale images. Our method is based on measuring Smoothness and Connectivity of neighbors of a specific pixel in image bitplanes. This method can produce a signed integer for each pixel capacity such that the sign shows the modification in pixel value must be increasing or decreasing. Comparison with the existing methods showed that the proposed method gave better estimation for capacity of data hiding in... 

This paper presents a new method for eye detection in a sequence of video pictures. We first find the face region using the features of skin's color, then the location of eye is determined by applying three cues on the detected face region. The first cue is the eye intensity and color, because the intensity of each eye region is relatively low and its color differs from the color of surrounding skin. The second cue is based on the position and size of eyes. The third cue is based on the convolution of proposed-eye-variance-filter with the eye-window candidates. After finding the precise location of eye, it will be tracked in a finite search window. We have evaluated and tested the presented... 

This study presents the utilisation of neural-network for bi-level image compression. In the proposed lossy compression method, the locations of pixels of image are applied to the inputs of a multilayer perceptron neural-network. The output of the network denotes the pixel intensity (0 or 1). The final weights of the trained neural-network are quantised, represented by a few bits, Huffman encoded and then stored as the compressed image. In the decompression phase, by applying the pixels locations to the trained network, the output determines the intensity. The results of experiments on more than 4000 different images indicate higher compression rate of the proposed structure compared with... 

This paper introduces a new method for the least significant bit (LSB) steganography in spatial domain. The new method, called one-third LSB embedding, reduces the probability of change per pixel to one-third without sacrificing the embedding capacity, as compared to previous LSB steganography techniques. This improvement results in a better imperceptibility and higher robustness against well known LSB detectors. Each bit of the message is carried through a function of three adjacent cover pixels and it is shown that no significant improvement is achieved by increasing the length of the pixel sequence employed. ©2009 IEEE  

In this paper, an advanced integrated pixel-level method based on the deep convolutional neural network (DCNN) approach named DeepLabv3+ is proposed for weak interlayers detection and quantification. Furthermore, a database containing 32,040 images of limestone, dolomite, loess clay, and red clay is established to verify this method. The proposed model is then trained, validated, and tested via feeding multiple weak interlayers. Moreover, robustness and adaptability of the proposed model are evaluated, and the weak interlayers are extracted. Compared with the fully convolutional network (FCN)-based method and traditional image techniques, the proposed model provides higher accuracy in terms... 

In this paper we propose a novel hierarchical Bayesian model for sparse regression problem to use in semi-supervised hyperspectral unmixing which assumes the signal recorded in each hyperspectral pixel is a linear combination of members of the spectral library contaminated by an additive Gaussian noise. To effectively utilizing the spatial correlation between neighboring pixels during the unmixing process, we exploit a Markov random field to simultaneously group pixels to clusters which are associated to regions with homogeneous mixtures in a natural scene. We assume Sparse fractional abundances of members of a cluster to be generated from an exponential distribution with the same rate... 

Similarity measure is an important key in image registration. Most traditional intensity-based similarity measures (e.g., SSD, CC, MI, and CR) assume stationary image and pixel by pixel independence. Hence, perfect image registration cannot be achieved especially in presence of spatially-varying intensity distortions and outlier objects that appear in one image but not in the other. Here, we suppose that non stationary intensity distortion (such as Bias field or Outlier) has sparse representation in transformation domain. Based on this as-sumption, the zero norm (ℓ0)of the residual image between two registered images in transform domain is introduced as a new similarity measure in presence... 

In this paper a new probabilistic view to the Visual Secret Sharing (VSS) is presented and as a result, a new VSS scheme is proposed. The proposed scheme is a threshold scheme with perfect black pixel reconstruction. We present the scheme by introducing two sets C0 and C1. We consider an unknown probability distribution on the elements of C0 and C 1. Perfectness provides some linear constraints on the probabilities of C0 and C1 elements. By solving the provided system of linear equations, the probability distributions are specified and we achieve a perfect VSS scheme. ©2008 IEEE  

We propose a new algorithm for object tracking in crowded video scenes by exploiting the properties of undecimated wavelet packet transform (UWPT) and interframe texture analysis. The algorithm is initialized by the user through specifying a region around the object of interest at the reference frame. Then, coefficients of the UWPT of the region are used to construct a feature vector (FV) for every pixel in that region. Optimal search for the best match is then performed by using the generated FVs inside an adaptive search window. Adaptation of the search window is achieved by interframe texture analysis to find the direction and speed of the object motion. This temporal texture analysis... 

This paper introduces a novel, cost effective visual cryptography scheme suitable for color image transmission over bandwidth constraint channels. Unlike previously proposed schemes, the solution offers perfect reconstruction while producing shares with size smaller than that of the input image. The maximum distance separable (MDS) code principle used in the design allows for the introduction of a flexible framework that compares favorably to competing solutions as it can be seen by examining the experimental results included in this paper. ©2008 IEEE  

Mesh-based motion estimation is an important tool for video coding especially with low bit rate. In this paper, a new method for interpolating pixel motion from adjacent mesh nodes with the ability of omitting independent nodes is proposed. By exploiting fuzzy rules to determine the association of pixel and neighboring nodes, the proposed interpolation can detach pixels from some nodes. Consequently, it can deal with those critical patches on objects boundary which their nodes do not belong to one object. Updating the membership functions of each rule with specified strategy makes the interpolation adaptive with non-stationary conditions of image sequences and decreases sensitivity to... 

Motion estimation is an important field of video processing. Conventional block motion estimation methods consider all pixels equally. In this paper a new edge sensitive algorithm is proposed in which distortion of ridgelet coefficients is calculated partially and compared over all candidates. Using an edge boosting distortion function, edge pixels will have more important role since they introduce larger ridgelet coefficients. Inherent ability of ridgelet in representing edges with various directions in multi scale, makes it a powerful tool for edge pursuit. Proposed partial ridgelet distortion search (PRDS) do not reject any candidate without assessment while performs relatively fast since... 

Multi-dimensional spatial analysis of image pixels have not been much investigated for the steganalysis of the LSB Steganographic methods. Pixel distribution based steganalysis methods could be thwarted by intelligently compensating statistical characteristics of image pixels, as reported in several papers. Simple LSB replacement methods have been improved by introducing smarter LSB embedding approaches, e.g. LSB matching and LSB+ methods, but they are basically the same in the sense of the LSB alteration. A new analytical method to detect LSB stego images is proposed in this paper. Our approach is based on the relative locations of image pixels that are essentially changed in an LSB... 

Fractional-pixel accuracy Motion Estimation (ME) has been shown to result in higher quality reconstructed image sequences in hybrid video coding systems. However, the higher quality is achieved by notably increased Motion Field (MF) bitrate and more complex computations. In this paper, new half-pixel block matching ME algorithms are proposed to improve the rate-distortion characteristics of low bitrate video communications. The proposed methods tend to decrease the required video bandwidth, while improving the motion compensation quality. The key idea is to put a deeper focus on the search origin of the ME process, based on center-bias characteristics of low bitrate video MFs. To employ the...