An iterative filtering algorithm based on signaling game idea

Mode filtering consists of modifying the domain point-pixel value of a template by the maximum of the probability density function of the image values contained within the template support, estimated by the local histogram. The limitations of mode filtering are addressed by feature selective templates utilized as structuring elements in morphological hit-miss transformations which preserve the edge intensity patterns potentially corrupted by the mode filtering operation. For nonunique modes, the central template pixel value is modified to the mode pixel value with the minimum absolute difference between these modes and the central template pixel. If the modes are symmetric about the central pixel value, producing equivalent minimum absolute differences, the template size is adaptively increased until a unique mode or a unique minimum absolute difference is obtained. Simulations are provided demonstrating the performance improvement of feature selective template mode filtering relative to other nonlinear image enhancement operations. >

Twenty years ago, Kay proposed an iterative filtering algorithm (IFA) for jointly estimating the frequencies of multiple complex sinusoids from noisy observations. IFA is based on the fact that the noiseless signal is an autoregressive (AR) process, so the frequency estimation problem can be reformulated as the problem of estimating the AR coefficients. By iterating the cycle of AR coefficient estimation and AR filtering, IFA provides a computationally simple procedure yet capable of accurate frequency estimation especially at low signal-to-noise ratio (SNR). However, the convergence of IFA has not been established beyond simulation and a very special case of a single frequency and infinite sample size. This article provides a statistical analysis of the algorithm and makes several important contributions. It shows that the poles of the AR filter must be reduced by an extra shrinkage parameter to accommodate poor initial values and avoid being trapped into false solutions. It also shows that the AR estimates in each iteration must be bias-corrected to produce a more accurate frequency estimator; a closed-form expression is provided for bias correction. Finally, it shows that for a sufficiently large sample size, the resulting algorithm, called new IFA (NIFA), converges to the desired fixed point, which constitutes a consistent frequency estimator. Numerical examples, including a real data example in radar applications, are provided to demonstrate the findings. It is shown in particular that the shrinkage parameter not only controls the estimation accuracy, but also determines the requirements of initial values. It is also shown that the proposed bias-correction method considerably improves the estimation accuracy, especially for high SNR.

Face recognition is a method which identifies a person based on the profile or features of the face of that person. LBPH (Local Binary Patterns Histogram) is a method to detect and recognize the face of a person. In LBP, first, some portion of an image which is in grayscale is taken as 3×3 window size and the pixel value of neighborhood is compared with the central pixel value and then the binary value is assigned which is then converted to a decimal value. LBP is then combined with histograms and so is called an LBPH algorithm. GPU (Graphics Processing Unit) is an electronic circuit which is more powerful than CPU (Central Processing Unit). In this paper, a frontal face and side profile face recognition using LBPH algorithm are implemented on GPU. The performance of the CPU and GPU are then compared.

The iterative extended Kalman filter (IEKF) algorithm is dependent on the previous data and estimation accuracy of the noise. The noise will result in filtering precision decrease and divergence. This paper puts forward an adaptive algorithm based on Levenberg-Marquardt method and two-factor for IEKF, which optimizes the iterative process by using Levenberg-Marquardt method, adjusts the prediction covariance by suboptimal fading factor. A simplified calculation method of suboptimal fading factor is also presented. The previous process noise is suppressed by exponential fading factor, and the observation noise is corrected by the residual and innovation, so that filter gain becomes larger, which improves the accuracy and robustness of the filter algorithm. The experiment results indicate the improved algorithm can effectively suppress the filter divergence. Compared with the classical extended Kalman filter (EKF), IEKF, and the modified IEKF, the filter accuracy and robustness of the improved algorithm are sufficient, and the algorithm can better meet the requirements of engineering applications.

This paper derives a new procedure for age classification of facial image based on the local region of facial image. The local region of facial image is extracted from a Significant Binary Pattern of Local Maximum Edge (SBPLME). The SBPLME is generated by calculating the absolute value of local difference between the average of local 3×3 sub window pixel values and its neighbors instead of the center pixel value. In the case of Local Maximum Edge Binary Pattern (LMEBP) calculating the absolute value of local difference between the center pixel value of local 3×3 sub window and its neighbors. The proposed SBPLME can generate 512 (0 to 511) different patterns. The present paper utilized Prominent LBP (PLBP) on the proposed SBPLME. The PLBP contains the significant patterns of Uniform LBP (ULBP) and Non Uniform LBP (NULBP). Thus the derived Significant PLBP of Local Maximum Edge (SPBPLME) becomes an efficient image classification and analysis, which will have a significant role in many areas. The novelty of the proposed SPBPLME method is, it has shown excellent age classification results by reducing the overall dimension, thus reducing the overall complexity.

The calculation of perspective projection is the essential part in either cone-beam projection or realistic volume rendering. In this paper, we propose a pixel-pyramid model to accurately calculate pixel values in flat-panel-based cone-beam reconstruction. This model faithfully portrays the divergent beam projection on a grilled planar detector. It consists of a pyramidal base (detector pixel facet) and an apex (X-ray source). Through spatial clipping by a flat-panel detector (FPD), a cone is tailored into a pyramid with the base at the FPD, called FPD pyramid, which is then spatially decomposed into pyramidal elements (PYREL). During cone-beam projection, a PYREL intersects the 3D object and produces a frustum, wherein the voxels are accumulated to produce a pixel value. In backprojection, the pixel value is imparted to the same contributive voxels. A voxel’s weight (contribution to a PYREL) is determined by spatial overlapping, which is a piecewise linear function. Through iterative algorithm, accurate cone beam reconstruction is achieved. Since the pyrel model portrays the perspective projection geometry and the voxel weight in a more natural fashion, it can be used for accurate calculation of perspective projection, especially in dealing with severe incomplete data. Simulations and comparisons are provided.

A Novel Texture Classification Scheme based on Completed Multiple Adaptive Threshold Patterns

HoMeNL: A Homogeneity Measure-Based NonLocal Filtering Framework for Detail-Enhanced (Pol)(In)SAR Image Denoising

Recognition of patterns in a person by using parts of the human body, such as on fingerprints, has been widely applied in life such as to perform absenteeism, tracking a criminal, system security and so on. Local Binary Pattern (LBP) algorithm is known as an algorithm that can describe local texture pattern in an area. LBP uses 8 scattered circular neighborhoods with center pixels centered. In a 3 x 3 pixel image, the binary value in the image center is compared with the surrounding value. The surrounding value will be 1 if the central pixel value is smaller, and is 0 if the central binary value is greater. A total of 78 data were used for this study where 26 data were using blue ink fingerprints, and 26 black ink data. After the fingerprint pattern data obtained then the image is scanned. After that the image in the crop to be 50 pixels x 50 pixels, so all the data becomes uniform. The algorithm used to make an introduction is the Manhattan Distance algorithm. Based on the test results of 26 test data with different color inks, the result obtained accuracy of 61.54%.

Stochastic encoding represents a value using the probability of ones in a random bit stream. Computation based on this encoding has good fault-tolerance and low hardware cost. However, one of its major issues is long processing time. We have to use a long enough bit stream to represent a value to guarantee that random fluctuations introduce only small errors to final computation results. For example, for most digital image processing algorithms, we need a 512-bit stream to represent an 8-bit pixel value stochastically to guarantee that the final computation error is less than 5%. To solve this issue, this paper proposes to share bits between adjacent bit streams to represent adjacent deterministic values. For example, in image processing applications, the bit stream which represents the current pixel value can share parts of the bits in the bit stream which represents the previous pixel value. We use an image contrast stretching algorithm to evaluate this method. Our experimental results show that the proposed methods can improve the performance by 90%.

A water beam imaging system (WBIS) was developed and used to verify dose distributions for intensity modulated radiotherapy (IMRT) using dynamic multileaf collimator (MLC). The WBIS consists of a water container, a scintillator screen, a CCD camera, and a portable personal computer. The scintillation image is captured by the CCD camera. The pixel value in this image indicates the dose value in the scintillation screen. The verification is performed by comparing the WBIS image achieved from the measurement with dose distribution from the IMRT plan. Because of light scattering in the WBIS the image is blurred. An iterative reconstruction algorithm is proposed to remove the blurring effect of light scattering. From the measured image of a 10 cm /spl times/10 cm X-ray beam and the simulation result of the dose distribution using the Monte Carlo method, the blur function can be achieved. Based on this function, the proposed algorithm is applied to reconstruct the true dose distribution for an IMRT plan from the measured WBIS image. The reconstructed dose distributions are compared with Monte Carlo simulation results. Reasonable agreement can be observed from the comparison. The proposed approach makes it possible to carry out real-time quality assurance tasks for IMRT dose verification.

Aiming at the problems of easy loss of GPS positioning signals in outdoor environments and inaccurate map construction and position drift of traditional SLAM algorithms in outdoor scenes, this paper proposes a 3D LiDAR and inertial guidance tightly coupled SLAM algorithm. Firstly, inertial measurement unit (IMU) forward propagation is used to predict the current position, then backward propagation is used to compensate the motion distortion in the LiDAR data, and the point cloud alignment residuals are constructed based on the GICP algorithm, and then the iterative error state Kalman filter (IESKF) algorithm is utilized to complete the fusion of the point cloud residuals and the a priori position obtained from the forward propagation of the IMU to complete the state updating, and then the front-end fusion odometer is constructed. Next, a sparse voxel near-neighbor structure, iVox-based method, is employed to select key frames and construct local maps, leveraging spatial information during frame-map matching. This approach reduces the computational time required for point cloud alignment. Finally, the proposed algorithm is validated in real-world scenarios and on the outdoor open-source dataset KITTI. It is compared against mainstream algorithms, including FAST-LIO2 and LIO-SAM. The results demonstrate that the proposed approach achieves lower cumulative error, higher localization accuracy, and improved visualization with greater robustness in outdoor environments.

Biometrics is the study of methods of recognizing humans based on their behavioral and physical characteristics or traits. Face recognition is one of the biometric modalities that received a great amount of attention from many researchers during the past few decades because of its potential applications in a variety of security domains. Face recognition however is not only concerned with recognizing human faces, but also with recognizing faces of non-biological entities or avatars. Fortunately, the need for secure and affordable virtual worlds is attracting the attention of many researchers who seek to find fast, automatic and reliable ways to identify virtual worlds’ avatars. In this work, I propose new techniques for recognizing avatar faces, which also can be applied to recognize human faces. Proposed methods are based mainly on a well-known and efficient local texture descriptor, Local Binary Pattern (LBP). I am applying different versions of LBP such as: Hierarchical Multi-scale Local Binary Patterns and Adaptive Local Binary Pattern with Directional Statistical Features in the wavelet space and discuss the effect of this application on the performance of each LBP version. In addition, I use a new version of LBP called Local Difference Pattern (LDP) with other well-known descriptors and classifiers to differentiate between human and avatar face images. The original LBP achieves high recognition rate if the tested images are pure but its performance gets worse if these images are corrupted by noise. To deal with this problem I propose a new definition to the original LBP in which the LBP descriptor will not threshold all the neighborhood pixel based on the central pixel value. A weight for each pixel in the neighborhood will be computed, a new value for each pixel will be calculated and then using simple statistical operations will be used to compute the new threshold, which will change automatically, based on the pixel’s values. This threshold can be applied with the original LBP or any other version of LBP and can be extended to work with Local Ternary Pattern (LTP) or any version of LTP to produce different versions of LTP for recognizing noisy avatar and human faces images.

The DM (Data Matrix) two-dimensional barcode images collected on metal parts with complex lighting conditions are prone to low contrast and local shadow, and the problem of uneven brightness leads to the loss of two-dimensional barcode information, which brings difficulties to the subsequent processing. However, the existing binarization algorithms will have the problems of too much noise and long segmentation time when dealing with this kind of images. To solve these problems, an improved local threshold algorithm is proposed to improve the anti-noise performance and computational efficiency. First of all, the algorithm model is improved to reduce the sensitivity to noise, and then the integral image algorithm is used to speed up the calculation of the local mean. In addition, the gray value of the local central pixel and the local mean are used to adjust the threshold adaptively according to the gray change characteristics of each pixel neighborhood, so as to further improve the segmentation efficiency. The improved local threshold algorithm and other five algorithms are tested on images with complex illumination conditions. The results show that the improved local threshold algorithm can not only effectively overcome the influence of complex illumination, has a good anti-noise effect, but also greatly shorten the time of image segmentation.

This paper proposes a novel approach in image compression based on Local Binary Pattern (LBP). LBP has already been used as a simple texture descriptor, labeling the image pixels by looking at the points surrounding a central point (usually on a 3×3 neighborhood) and examining whether these neighbors' color values are greater or less than the central point and accordingly assigning a binary value to the corresponding bit. The description of image's local pattern results in an eight-bit binary description, but in order to restore the image from such a LBP description, the value of each central pixel is also needed. These two pieces of information, i.e. the LBP description and the actual original value for each local neighborhood central pixel, are stored in a newly proposed Local Binary Compressed format, denoted .LBC, from which the image can be reconstructed by employing statistical methods, i.e. generating smaller or larger sets of random numbers to fill in the missing information within each local neighborhood, based on the LBP descriptor. Two statistical distributions were tested and, apart from the compression performance, a Structural Similarity Index Metric was used to evaluate the results.