PCA-Based Advanced Local Octa-Directional Pattern (ALODP-PCA): A Texture Feature Descriptor for Image Retrieval

This paper presents a novel feature descriptor termed principal component analysis (PCA)-based Advanced Local Octa-Directional Pattern (ALODP-PCA) for content-based image retrieval. The conventional approaches compare each pixel of an image with certain neighboring pixels providing discrete image information. The descriptor proposed in this work utilizes the local intensity of pixels in all eight directions of its neighborhood. The local octa-directional pattern results in two patterns, i.e., magnitude and directional, and each is quantized into a 40-bin histogram. A joint histogram is created by concatenating directional and magnitude histograms. To measure similarities between images, the Manhattan distance is used. Moreover, to maintain the computational cost, PCA is applied, which reduces the dimensionality. The proposed methodology is tested on a subset of a Multi-PIE face dataset. The dataset contains almost 800,000 images of over 300 people. These images carries different poses and have a wide range of facial expressions. Results were compared with state-of-the-art local patterns, namely, the local tri-directional pattern (LTriDP), local tetra directional pattern (LTetDP), and local ternary pattern (LTP). The results of the proposed model supersede the work of previously defined work in terms of precision, accuracy, and recall.

A Comparative Study of Image Descriptors in Recognizing Human Faces Supported by Distributed Platforms

Face recognition is one of the emergent technologies that has been used in many applications. It is a process of labeling pictures, especially those with human faces. One of the critical applications of face recognition is security monitoring, where captured images are compared to thousands, or even millions, of stored images. The problem occurs when different types of noise manipulate the captured images. This paper contributes to the body of knowledge by proposing an innovative framework for face recognition based on various descriptors, including the following: Color and Edge Directivity Descriptor (CEDD), Fuzzy Color and Texture Histogram Descriptor (FCTH), Color Histogram, Color Layout, Edge Histogram, Gabor, Hashing CEDD, Joint Composite Descriptor (JCD), Joint Histogram, Luminance Layout, Opponent Histogram, Pyramid of Gradient Histograms Descriptor (PHOG), Tamura. The proposed framework considers image set indexing and retrieval phases with multi-feature descriptors. The examined dataset contains 23,707 images of different genders and ages, ranging from 1 to 116 years old. The framework is extensively examined with different image filters such as random noise, rotation, cropping, glow, inversion, and grayscale. The indexer’s performance is measured based on a distributed environment based on sample size and multiprocessors as well as multithreads. Moreover, image retrieval performance is measured using three criteria: rank, score, and accuracy. The implemented framework was able to recognize the manipulated images using different descriptors with a high accuracy rate. The proposed innovative framework proves that image descriptors could be efficient in face recognition even with noise added to the images based on the outcomes. The concluded results are as follows: (a) the Edge Histogram could be best used with glow, gray, and inverted images; (b) the FCTH, Color Histogram, Color Layout, and Joint Histogram could be best used with cropped images; and (c) the CEDD could be best used with random noise and rotated images.

A Reconfigurable Memory based Fast VLSI Architecture for Histogram Computation

Histogram computation is the crucial task used in processing so many image guided applications like pattern recognition, image segmentation etc. Image registration is one of the fundamental techniques for pre-processing of the images. Registration is the process of overlaying multiple images to geometrically align them. In medical Image processing, the improper registration can have negative impact on the analysis of the image which influences the final diagnosis. The accurate result of image registration is obtained by matching of multimodal images. Mutual Information is one of the commonly used techniques to find the similarity measurement between multi-modal images. Measurement of similarity requires a computation of histogram of individual images and joint histogram between the images. The hardware implementation of histogram computation has advantages in terms of flexible design, low power consumption, high speed, less execution time than the software implementation. This paper proposed a parallel algorithm for histogram computation. A memory based pipeline architecture is designed for implementing the proposed algorithm. The hardware mapping of the algorithm on FPGA is proposed and simulating them using Xilinx software tools.

DblurDoseNet: A Deep Residual Learning Network for Voxel Radionuclide Dosimetry Compensating for SPECT Imaging Resolution

PurposeCurrent methods for patient specific voxel-level dosimetry in radionuclide therapy suffer from a trade-off between accuracy and computational efficiency. Monte Carlo (MC) radiation transport is considered as the gold standard but is computationally expensive, while faster dose voxel kernel (DVK) convolution can be sub-optimal in the presence of tissue heterogeneities. Furthermore, the accuracies of both these methods are limited by the spatial resolution of the reconstructed emission image. To overcome these limitations, this paper takes a novel approach of constructing a single deep convolutional neural network (CNN) named as DblurDoseNet that learns to produce dose-rate maps while compensating for the limited resolution of SPECT images.MethodsTo mitigate the effects of poor SPECT resolution and reconstruction artifacts on dosimetry, we trained our CNN using MC-generated dose-rate maps that directly corresponded to the true activity maps in virtual patient phantoms. We applied residual learning such that our CNN only learned the difference between the true dose-rate map and DVK dose-rate map with density scaling. The network consists of a depth feature extractor and a 2D U-Net, where the input was 11 slices (3.3 cm) of Lu-177 SPECT/CT images and the output was the dose-rate map corresponding to the center slice. In addition to phantoms, 42 SPECT/CT scans of patients who underwent Lu-177 DOTATATE therapy were also used for testing.ResultsIn test phantoms, the lesion/organ mean dose-rate error and the normalized root mean square error (NRMSE) relative to ground-truth for the CNN method was consistently lower than DVK and MC. In particular, for CNN compared to DVK/MC, the average improvement in mean dose error was 55%/53% and 66%/56%; and in NRMSE was 18%/17% and 10%/11% for lesion and kidney, respectively. Line profiles and dose-volume histograms demonstrated compensation for SPECT resolution effects in the CNN generated dose-rate maps. Noise, determined from multiple Poisson realizations, showed an average improvement of 21%/27% compared to DVK/MC. In patients, a high concordance was observed between CNN and MC in joint histogram analysis. The trained residual CNN took ~ 30 seconds on GPU to generate a (\(512\times 512\times 130\)) dose-rate map for a patient.ConclusionThe proposed CNN is well-suited for real-time patient-specific dosimetry for clinical treatment planning due to its demonstrated improvement in accuracy, resolution, noise and speed over the current gold-standard.

Pemanfaatan Histogram Equalization pada Local Tri Directional Pattern untuk Sistem Temu Kembali Citra

Local tri-directional pattern (LtriDP) is a method of extracting local intensity features from each pixel based on direction. However, this method has not been able to provide good performance in extracting features for image retrieval. One reason that makes image retrieval performance worse is the effect of lighting. Lighting can cause large variations between images. This study proposed utilization of Histogram Equalization (HE). Histogram equalization is a functional method of stretching gray degrees and expanding image contrast. This will make variations in the gray level of the original image can be controlled. There are several main stages in this study, firstly query image and image dataset will be preprocessed with histogram equalization. After that, the image is extracted by a tri-directional pattern and magnitude pattern are searched. A tri-directional pattern will produce two histograms, while a magnitude pattern produces one histogram. The three histograms are combined or joint histogram is performed. Histogram that has been joint is a feature vector. The feature vector will be calculated using a similarity measurement Canberra. After that, an image similar to the query image will be obtained. The experiment was conducted using 3 face datasets namely ORL, BERN, and YALE. The average recall value was 0.422 for the ORL dataset, 0.50 for the BERN dataset, and 0.63 for the YALE dataset. The evaluation show, the proposed method can be used as a process of improving the quality of image datasets in the image retrieval system. Keywords: Image retrieval system, Local tri-directional pattern, Streching Image, Histogram Equalization, Similarity Measurement Canberra. Abstrak Local tri-directional pattern (LtriDP) merupakan salah satu metode ekstraksi fitur intensitas lokal dari setiap piksel berdasarkan arah. Namun, metode ini belum mampu memberikan performa yang baik dalam mengekstrak fitur untuk temu kembali citra. Salah satu alasan yang membuat performa temu kembali citra tidak baik adalah pengaruh pencahayaan. Pencahayaan dapat menyebabkan variasi besar antar citra. Penelitian ini mengusulkan pemanfaatan Histogram Equalization (HE). HE merupakan metode fungsional dalam peregangan derajat keabuan dan memperluas kontras citra. Hal ini akan membuat variasi level keabuan dari citra asli dapat terkendali. Ada beberapa tahapan utama dalam penelitian ini, yang pertama citra query dan citra dataset akan terlebih dahulu di preprocessing dengan histogram equalization. Setelah itu, citra tersebut diekstrak fiturnya, dicari pola tri-directional dan pola magnitude. Pola tri-directional akan menghasilkan dua histogram, sedangkan pola magnitude menghasilkan satu histogram. Ketiga histogram tersebut kemudian disatukan atau dilakukan joint histogram. Histogram yang telah dijoint merupakan vektor fitur. Vektor fitur tersebut akan dihitung rankingnya menggunakan pengukuran jarak canberra. Setelah itu, akan didapatkan citra yang mirip dengan citra query. Uji coba dilakukan dengan menggunakan 3 dataset wajah yaitu ORL, BERN, dan YALE. Nilai rata-rata recall yang di dapatkan 0,422 untuk dataset ORL, 0,50 untuk dataset BERN, dan 0,63 untuk dataset YALE. Dari hasil evaluasi tersebut, dapat disimpulkan metode yang diusulkan dapat digunakan sebagai proses peningkatan kualitas dataset citra pada system temu kembali citra. Keywords: Sistem Temu Kembali Citra, Local tri-directional pattern, Peregangan Kontras, Histogram Equalization, Perhitungan Jarak Canberra.

Relationship Between Kendall's tau Correlation and Mutual Information

Mutual information (MI) can be viewed as a measure of multivariate association in a random vector. However, the estimation of MI is difficult since the estimation of the joint probability density function (PDF) of non Gaussian distributed data is a hard problem. Copula function is an appropriate tool for estimating MI since the joint probability density function ofrandom variables can be expressed as the product of the associated copula density function and marginal PDF’s. With a little search, we find that the proposed copulas-based mutual information is much more accurate than conventional methods such as the joint histogram and Parzen window-based MI. In this paper, by using the copulas-based method, we compute MI forsome family of bivariate distribution functions and study the relationship between Kendall’s tau correlation and MI of bivariate distributions. Finally, using a real dataset, we illustrate the efficiency of this approach.

Satellite-Based Assessment of Various Cloud Microphysics Schemes in Simulating Typhoon Hydrometeors

The accurate simulation of typhoon hydrometeors remains a challenge. This study attempts to evaluate the performances of five microphysics schemes (MPSs) in the Weather Research and Forecasting (WRF) model in simulating the supertyphoon Neoguri in July 2014. The observed microwave brightness temperature, as well as retrieved data from the microwave radiometer imager (MWRI) onboard Chinese FY-3B satellite, are used to test hydrometeor simulations. In particular, two MWRI radiance indices, including the emission index (EI) and scattering index (SI), are used to assess the performance of five MPSs in simulating liquid and frozen hydrometeors, respectively. Overall, the WRF model can well reproduce the overall pattern of typhoon-produced precipitation, albeit with slightly overestimated precipitation in the inner rainband and underestimated precipitation in the stratiform rainband. Moreover, ice water paths (IWPs) from all five MPS simulations are higher than those estimated from MWRI retrieval in most areas, and the spatial pattern and values of IWP for the National Severe Storms Laboratory double-moment MPS (NSSL) are much closer to those for MWRI. The NSSL scheme reproduces a more realistic joint histogram distribution of SI and EI than other MPSs do, relative to the observation. Besides, the nonlinear Lucas–Kanade optical flow approach has been used to reflect the horizontal distribution of hydrometeors in the typhoon. The results show that the simulated EI and SI from the five MPSs show a systematic southwest bias of approximately about 10∼20 km and significant intensity bias in the convection area. Further model sensitivity tests confirm that the NSSL scheme generates more realistic graupel and supercooled water close to the observations among all MPSs. The findings suggest that satellite measurements would be helpful to assess MPSs in numeric weather models, especially for hydrometeor distributions in the whole typhoon system.

Ex vivo evaluation of an atherosclerotic human coronary artery via histology and high-resolution hard X-ray tomography

Atherosclerotic arteries exhibit characteristic constrictions and substantial deviations from cylindrical shape. Therefore, determining the artery’s cross-section along the centerline is challenging, although high-resolution isotropic three-dimensional data are available. Herein, we apply high-resolution computed tomography in absorption and phase to a plaque-containing human artery post-mortem, through the course of the preparation stages for histology. We identify the impact of paraffin embedding and decalcification on the artery lumen. For automatic extraction of lumen’s cross-section along centerline we present a dedicated pipeline. Comparing fixated tissue before and after paraffin embedding gives rise to shape changes with lumen reduction to 50–80%. The histological slicing induces further deformations with respect to tomography. Data acquired after decalcification show debris unintentionally distributed within the vessel preventing the reliable automatic lumen segmentation. Comparing tomography of laboratory- and synchrotron-radiation-based X rays by means of joint histogram analysis leads us to conclude that advanced desktop tomography is capable of quantifying the artery’s lumen as an essential input for blood flow simulations. The results indicate that the most reliable lumen quantification is achieved by imaging the non-decalcified specimen fixed in formalin, using phase contrast modality and a dedicated processing pipeline. This study focusses on a methodology to quantitatively evaluate diseased artery segments post-mortem and provides unique structural parameters on the treatment-induced local shrinkage, which will be the basis of future studies on the flow in vessels affected by constrictions.