Whole Set Of Images Research Articles

Instagram public and private images as an element for the spatial monitoring of tourist behaviour in cities

PurposeThe purpose of this paper is to analyse whether publicly shared images on Instagram are representative of tourist behaviour in a destination. This aspect is crucial for destination image management, as it can influence the way tourists perceive the destination.Design/methodology/approachThe research compares three different factors: the route followed by a group of tourists, the whole set of photographs taken by them and the images that they made publicly available on Instagram. It relies on a field work done by a group of 122 tourists in Turin (Italy). At a qualitative level, the answers given by tourists to the motivations that led them to share some of the image are analysed.FindingsThe results showed how the spatial distribution of the images shared publicly on Instagram only partially coincides with the whole set of images taken by tourists.Practical implicationsIt is important to avoid basing marketing and management policies on just the places featured in publicly shared images. If not, there is a risk of taking decisions based on the behaviour of some, rather than all, tourists.Originality/valueMany papers claim to be based on the Instagram image as elements to study tourism. However, most of these papers only analyse public images. This fact can affect the results as there may be, for example, areas visited by tourists where photos are not taken. This paper therefore contributes to a better understanding of Instagram as a tool for the study of urban tourism.

Incremental PCA algorithm for fringe pattern demodulation

This work proposes a new algorithm for demodulating fringe patterns using principal component analysis (PCA). The algorithm is based on the incremental implantation of the singular value decomposition (SVD) technique for computing the principal values associated with a set of fringe patterns. Instead of processing an entire set of interferograms, the proposed algorithm proceeds in an incremental way, processing sequentially one (as minimum) interferogram at a given time. The advantages of this procedure are twofold. Firstly, it is not necessary to store the whole set of images in memory, and, secondly, by computing a phase quality parameter, it is possible to determine the minimum number of images necessary to accurately demodulate a given set of interferograms. The proposed algorithm has been tested for synthetic and experimental interferograms showing a good performance.

Frequency and anatomic distribution of magnetic resonance imaging lesions in the sacroiliac joints of spondyloarthritis and non-spondyloarthritis patients.

Background:Lesions detected by magnetic resonance imaging (MRI) of the sacroiliac joints are critical to the diagnosis of non-radiographic axial spondyloarthritis. However, inflammatory and structural lesions may be encountered in other conditions.Objectives:The objective of this study was to evaluate and compare the frequency and localization of inflammatory and structural lesions on MRIs of the sacroiliac joint of spondyloarthritis (SpA) and non-spondyloarthritis (non-SpA) patients.Design:This is a retrospective study including 200 patients, each having undergone an MRI of the sacroiliac joints.Methods:Two experienced readers evaluated the whole set of images to detect erosions, subchondral sclerosis, fatty lesions, bone marrow edema (BME) and ankylosis according to the definitions established by the ASAS MRI working group. We divided sacroiliac joints into five segments: upper, antero-middle, intermediate-middle, postero-middle and lower.Results:A total of 96 subjects with SpA (mean age 37.4 ± 11.8 years) and 104 without SpA (mean age 39.9 ± 11.6 years) were included. Of the 96 SpA patients, 65% had inflammatory buttock pain compared with 25% in the non-SpA group. BME was seen in 65% of SpA patients, mainly in the intermediate-middle segment, and in 20% of non-SpA patients, predominantly in the antero-middle segment. Subchondral sclerosis occurred in 44% of non-SpA patients, mostly in the antero-middle segment, and in 36% of SpA patients. Fatty lesions were present in 34% of SpA and in 21% of non-SpA patients. Erosions were seen in 25% of non-SpA and in 60% of SpA patients. BME and structural lesions were minimally observed in the postero-middle segment in non-SpA patients.Conclusion:Inflammatory and structural lesions were observed in all segments of the joint in SpA, mainly in the middle segments, while lesions predominantly affected the antero-middle segment in non-SpA, and were uncommon in the postero-middle segment.

WITHDRAWN: Categorising Apple Fruit Diseases Employing Hybrid Neural Clustering Classifier

Agricultural field plays a major role in development of India. The farmers are supplemented with additional profit by exporting fruits from India. The devastating problem in economic losses and production in agricultural industry is caused by diseases in fruits. Manual identification of defected fruit is time consuming and a difficult approach. Based on the pictures representing the symptoms of the fruit, Web-based system helps non experts also in the identification of the fruit disease. In this paper ten types of apple fruit diseases are classified by using the proposed Hybrid Neural Clustering (HNC) Classifier. The results of the feature extraction vector is supplied as the input to the classifier. There are two phases in this proposed method .The first phase is used to train the images by using the K Means Clustering to cluster the vector points and followed up with the second phase which is used to test by applying the Feed Forward Back propagation neural network (FFBP). The output received is the whole set of images that are classified into the various apple fruit diseases. The various diseases classified are Alternia Rot, Black Rot, Scab, Grey Mould, Bitter Rot, Blister Spot, Brown Rot, White Rot, Blue Mould, Powdery Mildew. The proposed method is evaluated by precision, recall and F Measure and it scored higher values than existing methods. Proposed method provides 98% of accuracy than other existing method.

Multi-view 3D face reconstruction with deep recurrent neural networks

Image-based 3D face reconstruction has great potential in different areas, such as facial recognition, facial analysis, and facial animation. Due to the variations in image quality, single-image-based 3D face reconstruction might not be sufficient to accurately reconstruct a 3D face. To overcome this limitation, multi-view 3D face reconstruction uses multiple images of the same subject and aggregates complementary information for better accuracy. Though appealing, there are multiple challenges in practice. Among these challenges, the most significant is the difficulty to establish coherent and accurate correspondence among a set of images, especially when these images are captured under unconstrained in-the-wild condition. This work proposes a method, Deep Recurrent 3D FAce Reconstruction (DRFAR), to solve the task of multi-view 3D face reconstruction using a subspace representation of the 3D facial shape and a deep recurrent neural network that consists of both a deep convolutional neural network (DCNN) and a recurrent neural network (RNN). The DCNN disentangles the facial identity and the facial expression components for each single image independently, while the RNN fuses identity-related features from the DCNN and aggregates the identity specific contextual information, or the identity signal, from the whole set of images to estimate the facial identity parameter, which is robust to variations in image quality and is consistent over the whole set of images. Experimental results indicate significant improvement over state-of-the-art in both the accuracy and the consistency of 3D face reconstruction. Moreover, face recognition results on IJB-A with the UR2D face recognition pipeline indicate that, compared to single-view 3D face reconstruction, the proposed multi-view 3D face reconstruction algorithm can improve the face identification accuracy of UR2D by two percentage points in Rank-1 identification rate.

Utility of 8 h and time decay-corrected acquisition scintigraphy with in-vitro labeled white blood cells for the diagnosis of osteoarticular infection.

Except in the spine, labeled white-blood cell scintigraphy (WBCS) with image acquisition up to 24 h is the nuclear medicine test of choice for diagnosing osteoarticular infection. However, distinguishing between inflammation and infection is a challenge. The first aim of this study was to verify earlier research studies that used 4 and 24 h time decay-corrected acquisition (TDCA) to differentiate infection from inflammation. The second aim was to analyze whether 8 h acquisition (1-day protocol) yielded similar results as 20-24 h acquisition. This was an observational study of 94 patients (22-86 years, 52 women) with suspected osteoarticular infection referred to nuclear medicine to confirm infection. WBCS and TDCA images were obtained at 30 min, 4 h, and 8 h after injection of the labeled leukocytes, with collection times of 5, 8, and 12 min, respectively. Scintigrams were classified into three protocols: protocol 1: experts read only 30 min and 4 h images; protocol 2: experts read the whole set of images (30 min, 4 h, and 8 h) with different pixel intensities (each image normalized to its own maximum activity); protocol 3: experts read the whole set of images with the same pixel intensity. Sensitivity, specificity, positive and negative predictive values, and accuracy were calculated. In patients with orthopedic implants, the interobserver reproducibility for visual analysis was calculated using the κ index. Infection was confirmed in 26 cases. Sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and κ results were as follows: protocol 1: 92.3, 50.0, 41.4, 94.4, 61.7%, 0.79; protocol 2: 92.3, 94.1, 85.7, 97.0, 93.6%, 0.80; protocol 3: 96.2, 97.1, 92.6, 98.5, 96.8%, 0.77. TDCA acquisition of WBCS at 8 h (1-day protocol) enables a faster diagnosis than 24 h acquisition. The use of TDCA with the same pixel intensity in all images enables an accurate diagnostic of osteoarticular infection, with a considerable interobserver agreement for all protocols.

HIERARCHICAL STRUCTURE FROM MOTION COMBINING GLOBAL IMAGE ORIENTATION AND STRUCTURELESS BUNDLE ADJUSTMENT

Abstract. Global image orientation techniques aim at estimating camera rotations and positions for a whole set of images simultaneously. One of the main arguments for these procedures is an improved robustness against drifting of camera stations in comparison to more classical sequential approaches. Usually, the process consists of computation of absolute rotations and, in a second step, absolute positions for the cameras. Either the first or both steps rely on the network of transformations arising from relative orientations between cameras. Therefore, the quality of the obtained absolute results is influenced by tensions in the network. These may e.g. be induced by insufficient knowledge of the intrinsic camera parameters. Another reason can be found in local weaknesses of image connectivity. We apply a hierarchical approach with intermediate bundle adjustment to reduce these effects. We adopt efficient global techniques which register image triplets based on fixed absolute camera rotations and scaled relative camera translations but do not involve scene structure elements in the fusion step. Our variant employs submodels of arbitrary size, orientation and scale, by computing relative rotations and scales between - and subsequently absolute rotations and scales for - submodels and is applied hierarchically. Furthermore we substitute classical bundle adjustment by a structureless approach based on epipolar geometry and augmented with a scale consistency constraint.

Parallel Algorithms for Effective Correspondence Problem Solution in Computer Vision

We propose new parallel algorithms for correspondence problem solution in computer vision. We develop an industrial photogrammetric system that uses artificial retroreflective targets that are photometrically identical. Therefore, we cannot use traditional descriptor-based point matching methods, such as SIFT, SURF etc. Instead, we use epipolar geometry constraints for finding potential point correspondences between images. In this paper, we propose new effective graph-based algorithms for finding point correspondences across the whole set of images (in contrast to traditional methods that use 2-4 images for point matching). We give an exact problem solution via superclique and show that this approach cannot be used for real tasks due to computational complexity. We propose a new effective parallel algorithm that builds the graph from epipolar constraints, as well as a new fast parallel heuristic clique finding algorithm. We use an iterative scheme (with backprojection of the points, filtering of outliers and bundle adjustment of point coordinates and cameras’ positions) to obtain an exact correspondence problem solution. This scheme allows using heuristic clique finding algorithm at each iteration. The proposed architecture of the system offers a significant advantage in time. Newly proposed algorithms have been implemented in code; their performance has been estimated. We also investigate their impact on the effectiveness of the photogrammetric system that is currently under development and experimentally prove algorithms’ efficiency.

Multicriteria classification method for dimensionality reduction adapted to hyperspectral images

Due to the incredible growth of high dimensional datasets, we address the problem of unsupervised methods sensitive to undergoing different variations, such as noise degradation, and to preserving rare information. Therefore, researchers nowadays are forced to develop techniques to meet the needed requirements. In this work, we introduce a dimensionality reduction method that focuses on the multiobjectives of multiple images taken from multiple frequency bands, which form a hyperspectral image. The multicriteria classification algorithm technique compares and classifies these images based on multiple similarity criteria, which allows the selection of particular images from the whole set of images. The selected images are the ones chosen to represent the original set of data while respecting certain quality thresholds. Knowing that the number of images in a hyperspectral image signifies its dimension, choosing a smaller number of images to represent the data leads to dimensionality reduction. Also, results of tests of the developed algorithm on multiple hyperspectral image samples are shown. A comparative study later on will show the advantages of this technique compared to other common methods used in the field of dimensionality reduction.

Adaptive Spot Detection With Optimal Scale Selection in Fluorescence Microscopy Images.

Accurately detecting subcellular particles in fluorescence microscopy is of primary interest for further quantitative analysis such as counting, tracking, or classification. Our primary goal is to segment vesicles likely to share nearly the same size in fluorescence microscopy images. Our method termed adaptive thresholding of Laplacian of Gaussian (LoG) images with autoselected scale (ATLAS) automatically selects the optimal scale corresponding to the most frequent spot size in the image. Four criteria are proposed and compared to determine the optimal scale in a scale-space framework. Then, the segmentation stage amounts to thresholding the LoG of the intensity image. In contrast to other methods, the threshold is locally adapted given a probability of false alarm (PFA) specified by the user for the whole set of images to be processed. The local threshold is automatically derived from the PFA value and local image statistics estimated in a window whose size is not a critical parameter. We also propose a new data set for benchmarking, consisting of six collections of one hundred images each, which exploits backgrounds extracted from real microscopy images. We have carried out an extensive comparative evaluation on several data sets with ground-truth, which demonstrates that ATLAS outperforms existing methods. ATLAS does not need any fine parameter tuning and requires very low computation time. Convincing results are also reported on real total internal reflection fluorescence microscopy images.

SU‐FF‐T‐88: Image Registration for Real‐Time Dose Calculation of Adaptive Radiation Therapy

Purpose: To enhance the speed and accuracy of certain treatment plan in radiation therapy, a real‐time dose calculation strategy is performed based on Fast Fourier Transformation (FFT) and CT image registration. For this purpose, different registration methods are achieved by using Insight Segmentation and Registration Toolkit (ITK) in MATLAB. Method and materials: Our group developed an efficient method for kernel calculation for pencil beam convolution‐superposition based on FFT. In this paper, different image registration methods are developed in MATLAB by using ITK, such as rigid 3D, similarity 3D, Demons deformation and BSpline based Free Form. As ITK is written and used in a C++ template, smart pointer and generic structure, an open source wrapper MATITK is modified to facilitate the use of its image processing capabilities while working in the high‐level environment of MATLAB. The image registration methods are demonstrated on head and neck (H&N) CT image sets of 6 patients. The results of evaluation and visualization for the registration are presented. Results and discussion: Transformation similarity and computation complexity as evaluation criteria are compared between rigid and deformable registration methods. Judging from the mutual information histogram and mean square metric, the deformable registration will get more improvement on similarity than the rigid registration. Checkerboard composites of the reference and registered images are displayed. The computational time of deformable registration are 10∼20 times longer with the whole set of images because of the pre‐processing, complicated transformation matrix and more parameters. A similarity3D registration method is adopted for the image registration of H&N radiation therapy to trade off between registration effect and the time consumption. Conclusion: The image registration can greatly contribute to the automatic contour adjustment and dose deformation for adaptive radiation therapy. A GUI program will be designed for the customized deformation region to speed up deformable registration.

A Novel VLSI Architecture of Hybrid Image Compression Model based on Reversible Blockade Transform

Image compression can improve the performance of the digital systems by reducing time and cost in image storage and transmission without significant reduction of the image quality. Furthermore, the discrete cosine transform has emerged as the new state-of-the art standard for image compression. In this paper, a hybrid image compression technique based on reversible blockade transform coding is proposed. The technique, implemented over regions of interest (ROIs), is based on selection of the coefficients that belong to different transforms, depending on the coefficients is proposed. This method allows: (1) codification of multiple kernals at various degrees of interest, (2) arbitrary shaped spectrum,and (3) flexible adjustment of the compression quality of the image and the background. No standard modification for JPEG2000 decoder was required. The method was applied over different types of images. Results show a better performance for the selected regions, when image coding methods were employed for the whole set of images. We believe that this method is an excellent tool for future image compression research, mainly on images where image coding can be of interest, such as the medical imaging modalities and several multimedia applications. Finally VLSI implementation of proposed method is shown. It is also shown that the kernal of Hartley and Cosine transform gives the better performance than any other model. Keywords—VLSI, Discrete Cosine Transform, JPEG, Hartley transform, Radon Transform

Image compression: Maxshift ROI encoding options in JPEG2000

Image compression can improve the performance of the digital systems by reducing time and cost in image storage and transmission without significant reduction of the image quality. Furthermore, the JPEG2000 has emerged as the new state-of-the art standard for image compression. In this paper, a Selective Coefficient Mask Shift (SCMShift) coding method is proposed. The technique, implemented over regions of interest (ROIs), is based on shifting the wavelet coefficients that belong to different subbands, depending on the coefficients relative to the original image. This method allows: (1) codification of multiple ROIs at various degrees of interest, (2) arbitrary shaped ROI coding, and (3) flexible adjustment of the compression quality of the ROI and the background. No standard modification for JPEG200 decoder was required. The method was applied over different types of images. Results show a better performance for the selected regions, when ROI coding methods were employed for the whole set of images. We believe that this method is an excellent tool for future image compression research, mainly on images where ROI coding can be of interest, such as the medical imaging modalities and several multimedia applications.

Dodecameric Structure of the Small Heat Shock Protein Acr1 from Mycobacterium tuberculosis

Small heat shock proteins are a ubiquitous and diverse family of stress proteins that have in common an alpha-crystallin domain. Mycobacterium tuberculosis has two small heat shock proteins, Acr1 (alpha-crystallin-related protein 1, or Hsp16.3/16-kDa antigen) and Acr2 (HrpA), both of which are highly expressed under different stress conditions. Small heat shock proteins form large oligomeric assemblies and are commonly polydisperse. Nanoelectrospray mass spectrometry showed that Acr2 formed a range of oligomers composed of dimers and tetramers, whereas Acr1 was a dodecamer. Electron microscopy of Acr2 showed a variety of particle sizes. Using three-dimensional analysis of negative stain electron microscope images, we have shown that Acr1 forms a tetrahedral assembly with 12 polypeptide chains. The atomic structure of a related alpha-crystallin domain dimer was docked into the density to build a molecular structure of the dodecameric Acr1 complex. Along with the differential regulation of these two proteins, the differences in their quaternary structures demonstrated here supports their distinct functional roles.

Interactive segmentation of masses in digitized mammograms: observer variability and discriminative applications

This work presents a study on interactive segmentation of breast masses in digitized mammograms and on the analysis of shape parameters: compactness ( C), contour moment-based shape parameter ( M), and Fourier shape parameter ( F). Digitized mammograms were selected for this study from The University of South Florida Digital Database for Screening Mammography (DDSM), containing 94 masses (46 benign and 48 malignant) with different types of margins. Two observers participated in this study. Each observer performed the same procedure of segmentation with the whole set of images three times in three independent sessions. Our results show a high dependence of M on the shape of the segmented image and more sensitivity of C and F to contour irregularities. Intra- and inter-observer differences in human segmentation, measured by means of the shape parameters, are shown. Each single shape parameter allows differentiating between malignant and benign masses. When results from both observers are used together, the degree of discrimination is higher than when they are used independently. We have demonstrated also various degrees of discrimination for other characteristics of the masses such as assessment, margins and shape. As a conclusion, the use of several manually outlined contours as a gold standard for CAD systems can improve performance.