No-reference image quality assessment using interval type 2 fuzzy sets

Image quality assessment of distorted or decompressed images without any reference to the original image is challenging from computational point of view. Quality of an image is best judged by human observers without any reference image, and evaluated using subjective measures. The paper aims at designing a generic no-reference image quality assessment (NR-IQA) method by incorporating human visual perception in assigning quality class labels to the images. Using fuzzy logic approach, we consider information theoretic entropies of visually salient regions of images as features and assess quality of the images using linguistic values. The features are transformed into fuzzy feature space by designing an algorithm based on interval type-2 (IT2) fuzzy sets. The algorithm measures uncertainty present in the input–output feature space to predict image quality accurately as close to human observations. We have taken a set of training images belonging to five different pre-assigned quality class labels for calculating foot print of uncertainty (FOU) corresponding to each class. To assess the quality class label of the test images, maximum of T-conorm applied on the lower and upper membership functions of the test images belonging to different classes is calculated. Our proposed image quality metric is compared with other no-reference quality metrics demonstrating more accurate results and compatible with subjective mean opinion score metric.     

无参考屏幕内容图像质量评价

随着多客户端交互多媒体应用的快速发展,屏幕内容图像（Screen content image,SCI）的分发和处理与日俱增.图像质量评价课题的研究是其它许多应用的基础,至今图像质量评价课题研究的重点是传统自然图像,因此针对屏幕图像质量评价的研究就变的非常迫切和必要.客观图像质量评价算法的提出的基础建立在标准图像质量评价数据库上.本文首先构建了一个大规模的屏幕内容图像质量评价数据库（Immersive Media Laboratory screen content image quality database,IML-SCIQD）.IML-SCIQD数据库包含参考图像25张以及经过10种失真处理的1250张失真图像.以建立的IML-SCIQD数据库为基础,考虑到屏幕内容图像图像区域和文本区域的视觉感知差异,在基于自然场景统计的无参考方法的启发下,本文提出了针对屏幕内容图像的无参考评价算法（Natural Scene Statistics based No Reference Screen Content Image Quality Assessment metric,NSNRS）.NSNRS算法首先分别计算图像区域和文本区域的质量分数,再将这两个区域的质量分数结合起来得到整幅失真图像的质量分数.该算法与其它12种经典的客观评价算法,包括全参考算法,部分参考算法与无参考算法,在IML-SCIQD数据库和SIQAD数据库上进行了性能测试和对比,结果表明本文提出的算法优于经典的的无参考评价算法;就整个数据库而言,本文提出的算法可以达到与全参考方法相当的性能.     

A new standard method of subjective assessment of barely visible image artifacts and a new public database

Assessing image quality is an important aspect of developing new display technology. A particularly challenging assessment is determining whether a bitwise lossy operation is visually lossless. We define “visually lossless” and describe a new standard for a subjective procedure to assess whether the image quality meets these criteria. Assessments are made between a reference image and temporally interleaved reference and test images using a forced‐choice procedure. In extensive testing, we have validated that this method is suitable for discriminating between subtle differences in image rendering and is free of observer bias or criteria variability. The results of these tests demonstrate the efficacy of using as few as five randomly chosen observers. We have found that the subjective testing is more reliable than several widely available image quality metrics. As part of this work, we release a database of nearly 0.25 million subjective responses collected from 35 observers to 18 different images. The study uses a largely within‐subjects design and tested observers from two viewing distances. We encourage the use of this dataset in future research to refine objective image quality metrics to improve predictability of subtle but potentially visible compression‐induced image impairments.     

面向水下图像的质量评价方法

目的 针对目前水下图像质量评价方法少和现有方法存在局限性等问题,提出一种无参考并且无需手工设计特征的水下图像质量评价方法。方法 提出的水下图像质量评价方法将深度学习网络框架与随机森林回归模型相结合,首先采用深度神经网络提取水下图像的特征;然后使用提取的特征和标定的水下图像质量分数训练回归模型;最终,利用训练好的回归模型预测水下图像的质量。结果 在本文收集的水下图像数据集和水下图像清晰化算法处理结果上评测本文方法,并与多种质量评价方法进行比较,其中包括预测结果与主观质量分数比较、水下图像清晰化结果评测比较、预测结果与主观质量分数相关性比较、鲁棒性比较等。主观实验结果表明本文的评价方法可以相对准确地给出符合人类视觉感知的水下图像质量分数,并且具有更好的鲁棒性。定量实验结果表明本文方法与其他方法相比,预测的图像质量分数与主观分数具有更高的相关性。结论 提出的水下图像质量评价方法无需参考图像,省去了手工设计的特征,充分利用了深度学习网络的学习和表征能力。本文方法的准确性较好,普适性和鲁棒性较高,预测的质量分数与人类视觉感知具有较高的一致性。本方法适用于原始的水下图像和水下图像清晰化算法的处理结果。     

融合注意力机制和高效网络的糖尿病视网膜病变识别与分类

目的 糖尿病视网膜病变（diabetic retinopathy,DR）是一种病发率和致盲率都很高的糖尿病并发症。临床中,由于视网膜图像不同等级之间差异性小以及临床医生经验的不同,会出现误诊、漏诊等情况,目前基于人工DR的诊断分类性能差且耗时费力。基于此,本文提出一种融合注意力机制（attention mechanism）和高效率网络（high-efficiency network,EfficientNet）的DR影像自动分类识别方法,以此达到对病变类型的精确诊断。方法 针对实验中DR数据集存在的问题,进行剔除、去噪、扩增和归一化等处理;利用EfficientNet进行特征提取,采用迁移学习的策略用DR的数据集对EfficientNet进行学习与训练,提取深度特征。为了解决病变之间差异小的问题,防止网络对糖尿病视网膜图像的特征学习时出现错分等情况,在EfficientNet输出结果上加入注意力机制;根据网络提取的特征在深度分类器中进行分类,将视网膜图像按等级进行五分类。结果 本文方法的分类精度、敏感性、特异性和二次加权（kappa）值分别为97.2%、95.6%、98.7%和0.84,具有较好的分类性能及鲁棒性。结论 基于融合注意力机制的高效率网络（attention EfficientNet,A-EfficientNet）的DR分类算法有效地提高了DR筛查效率,解决了人工分类的手动提取特征的局限性,在临床上对医生诊断起到了辅助作用,能更有效地防治此类恶性眼疾造成严重视力损伤、甚至失明。     

From omnidirectional images to hierarchical localization

We propose a new vision-based method for global robot localization using an omnidirectional camera. Topological and metric localization information are combined in an efficient, hierarchical process, with each step being more complex and accurate than the previous one but evaluating fewer images. This allows us to work with large reference image sets in a reasonable amount of time. Simultaneously, thanks to the use of 1D three-view geometry, accurate metric localization can be achieved based on just a small number of nearby reference images. Owing to the wide baseline features used, the method deals well with illumination changes and occlusions, while keeping the computational load small. The simplicity of the radial line features used speeds up the process without affecting the accuracy too much. We show experiments with two omnidirectional image data sets to evaluate the performance of the method and compare the results using the proposed radial lines with results from state-of-the-art wide-baseline matching techniques.     

基于水平集的牙齿CT图像分割技术

牙齿的计算机断层扫描(CT)图像中存在边界模糊、相邻牙齿粘连等情况,且拓扑结构较为复杂,要实现准确的牙齿分割非常困难。对传统的牙齿CT图像分割方法,特别是近年来用于牙齿分割的水平集方法进行介绍,对其水平集函数中各能量项进行研究,并通过对比实验体现水平集方法的优越性。基于水平集的牙齿CT图像分割方法中水平集函数的能量项主要包括:竞争能量项、梯度能量项、形状约束能量项、全局先验灰度能量项、局部灰度能量项。实验结果表明基于混合模型的水平集方法分割效果最佳,切牙与磨牙分割准确率分别为88.92%和92.34%,相比自适应阈值和传统水平集方法,分割准确率总体提升10%以上。在综合利用图像信息和先验知识的基础上,通过对水平集函数中能量项进行优化和创新,有望进一步提高分割的准确率。     

多失真混杂的无人机图像质量盲评价

目的 面向多失真混杂的图像质量盲评价问题目前仍然是计算机视觉领域具有挑战性的工作之一,无人机图像受成像条件影响混杂多类失真,图像质量的准确评价是其效能发挥的关键环节。为此,引入并改进了基于自然场景统计的距离度量评价模型,提出多失真混杂的无人机图像质量盲评价方法。方法 从图像的结构性、信息完整性和颜色性3个不同的角度研究并提取了与无人机图像质量敏感的特征因子集;以实拍标准测绘图像库为原始图像获得MVG特性参数作为度量基准解决了盲评价中缺乏训练集的问题;构建了以实飞图像为样本的无人机图像质量数据库（UAV image set）,为相关问题的研究提供数据集和评价参考。结果 针对所构建的数据库,本文算法在主客观一致性、算法运行时间上与其他算法进行了对比实验。相比较其他经典算法,本文算法的主客观一致性较高,达到了0.8以上,运行时间较快,过到1.2 s。此外本文还给出了块大小对算法影响以及单特征对图像的评价结果,证明算法选择的图像块大小和图像特征符合质量评价的需要。结论 针对无人机图像所包含的多失真构建质量评价综合模型,该模型可满足无人机图像质量需求。     

Leveraging Retinal Fundus Images with Deep Learning for Diabetic Retinopathy Grading and Classification

Recently, there has been a considerable rise in the number of diabetic patients suffering from diabetic retinopathy (DR). DR is one of the most chronic diseases and makes the key cause of vision loss in middle-aged people in the developed world. Initial detection of DR becomes necessary for decreasing the disease severity by making use of retinal fundus images. This article introduces a Deep Learning Enabled Large Scale Healthcare Decision Making for Diabetic Retinopathy (DLLSHDM-DR) on Retinal Fundus Images. The proposed DLLSHDM-DR technique intends to assist physicians with the DR decision-making method. In the DLLSHDM-DR technique, image preprocessing is initially performed to improve the quality of the fundus image. Besides, the DLLSHDM-DR applies HybridNet for producing a collection of feature vectors. For retinal image classification, the DLLSHDM-DR technique exploits the Emperor Penguin Optimizer (EPO) with a Deep Recurrent Neural Network (DRNN). The application of the EPO algorithm assists in the optimal adjustment of the hyperparameters related to the DRNN model for DR detection showing the novelty of our work. To assuring the improved performance of the DLLSHDM-DR model, a wide range of experiments was tested on the EyePACS dataset. The comparison outcomes assured the better performance of the DLLSHDM-DR approach over other DL models.     

Automated accurate registration method between UAV image and Google satellite map

Because Unmanned Aerial Vehicle (UAV) image exhibits low positioning accuracy, the accurate registration of the image is required. Since the viewpoint direction, capturing time and shoot height are considerably different between the UAV image and google satellite map, the existing methods cannot match two images accurately. For the registration between the UAV image and google satellite map, a full-automated image registration method was proposed based on deep convolution feature. Such method consists of five steps: automatically reference images downloading, uniform key point extraction, deep convolution features computation, accurately feature matching and image registration. The reference image was downloaded from google map service according to the approximate location and region of the UAV image. The deep convolution feature was extracted using the pre-trained VGG16 model. Finally, many experiments were performed to verify the efficiency of the proposed method, and the results demonstrate that the proposed method is more effective and robust than the existing method.

     

Illumination invariant face recognition

The appearance of a face will vary drastically when the illumination changes. Variations in lighting conditions make face recognition an even more challenging and difficult task. In this paper, we propose a novel approach to handle the illumination problem. Our method can restore a face image captured under arbitrary lighting conditions to one with frontal illumination by using a ratio-image between the face image and a reference face image, both of which are blurred by a Gaussian filter. An iterative algorithm is then used to update the reference image, which is reconstructed from the restored image by means of principal component analysis (PCA), in order to obtain a visually better restored image. Image processing techniques are also used to improve the quality of the restored image. To evaluate the performance of our algorithm, restored images with frontal illumination are used for face recognition by means of PCA. Experimental results demonstrate that face recognition using our method can achieve a higher recognition rate based on the Yale B database and the Yale database. Our algorithm has several advantages over other previous algorithms: (1) it does not need to estimate the face surface normals and the light source directions, (2) it does not need many images captured under different lighting conditions for each person, nor a set of bootstrap images that includes many images with different illuminations, and (3) it does not need to detect accurate positions of some facial feature points or to warp the image for alignment, etc.     

Fingerprint image mosaicking by recursive ridge mapping.

To obtain a large fingerprint image from several small partial images, mosaicking of fingerprint images has been recently researched. However, existing approaches cannot provide accurate transformations for mosaics when it comes to aligning images because of the plastic distortion that may occur due to the nonuniform contact between a finger and a sensor or the deficiency of the correspondences in the images. In this paper, we propose a new scheme for mosaicking fingerprint images, which iteratively matches ridges to overcome the deficiency of the correspondences and compensates for the amount of plastic distortion between two partial images by using a thin-plate spline model. The proposed method also effectively eliminates erroneous correspondences and decides how well the transformation is estimated by calculating the registration error with a normalized distance map. The proposed method consists of three phases: feature extraction, transform estimation, and mosaicking. Transform is initially estimated with matched minutia and the ridges attached to them. Unpaired ridges in the overlapping area between two images are iteratively matched by minimizing the registration error, which consists of the ridge matching error and the inverse consistency error. During the estimation, erroneous correspondences are eliminated by considering the geometric relationship between the correspondences and checking if the registration error is minimized or not. In our experiments, the proposed method was compared with three existing methods in terms of registration accuracy, image quality, minutia extraction rate, processing time, reject to fuse rate, and verification performance. The average registration error of the proposed method was less than three pixels, and the maximum error was not more than seven pixels. In a verification test, the equal error rate was reduced from 10% to 2.7% when five images were combined by our proposed method. The proposed method was superior to other compared methods in terms of registration accuracy, image quality, minutia extraction rate, and verification.     

Simple methods for segmentation and measurement of diabetic retinopathy lesions in retinal fundus images

Diabetic retinopathy (DR) is one of the most important complications of diabetes mellitus, which causes serious damages in the retina, consequently visual loss and sometimes blindness if necessary medical treatment is not applied on time. One of the difficulties in this illness is that the patient with diabetes mellitus requires a continuous screening for early detection. So far, numerous methods have been proposed by researchers to automate the detection process of DR in retinal fundus images. In this paper, we developed an alternative simple approach to detect DR. This method was built on the inverse segmentation method, which we suggested before to detect Age Related Macular Degeneration (ARMDs). Background image approach along with inverse segmentation is employed to measure and follow up the degenerations in retinal fundus images. Direct segmentation techniques generate unsatisfactory results in some cases. This is because of the fact that the texture of unhealthy areas such as DR is not homogenous. The inverse method is proposed to exploit the homogeneity of healthy areas rather than dealing with varying structure of unhealthy areas for segmenting bright lesions (hard exudates and cotton wool spots). On the other hand, the background image, dividing the retinal image into high and low intensity areas, is exploited in segmentation of hard exudates and cotton wool spots, and microaneurysms (MAs) and hemorrhages (HEMs), separately. Therefore, a complete segmentation system is developed for segmenting DR, including hard exudates, cotton wool spots, MAs, and HEMs. This application is able to measure total changes across the whole retinal image. Hence, retinal images that belong to the same patients are examined in order to monitor the trend of the illness. To make a comparison with other methods, a Na?ve Bayes method is applied for segmentation of DR. The performance of the system, tested on different data sets including various qualities of retinal fundus images, is over 95% in detection of the optic disc (OD), and 90% in segmentation of the DR.     

Defect detection in patterned wafers using anisotropic kernels

Wafer defect detection often relies on accurate image registration of source and reference images obtained from neighboring dies. Unfortunately, perfect registration is generally impossible, due to pattern variations between the source and reference images. In this paper, we propose a defect detection procedure, which avoids image registration and is robust to pattern variations. The proposed method is based on anisotropic kernel reconstruction of the source image using the reference image. The source and reference images are mapped into a feature space, where every feature with origin in the source image is estimated by a weighted sum of neighboring features from the reference image. The set of neighboring features is determined according to the spatial neighborhood in the original image space, and the weights are calculated from exponential distance similarity function. We show that features originating from defect regions are not reconstructible from the reference image, and hence can be identified. The performance of the proposed algorithm is evaluated and its advantage is demonstrated compared to using an anomaly detection algorithm.     

Automated processing of shoeprint images based on the Fourier transform for use in forensic science

The development of a system for automatically sorting a database of shoeprint images based on the outsole pattern in response to a reference shoeprint image is presented. The database images are sorted so that those from the same pattern group as the reference shoeprint are likely to be at the start of the list. A database of 476 complete shoeprint images belonging to 140 pattern groups was established with each group containing two or more examples. A panel of human observers performed the grouping of the images into pattern categories. Tests of the system using the database showed that the first-ranked database image belongs to the same pattern category as the reference image 65 percent of the time and that a correct match appears within the first 5 percent of the sorted images 87 percent of the time. The system has translational and rotational invariance so that the spatial positioning of the reference shoeprint images does not have to correspond with the spatial positioning of the shoeprint images of the database. The performance of the system for matching partial-prints was also determined.     

Image quality evaluation based on recognition times for fast image browsing applications

Mean squared error (MSE) and peak signal-to-noise-ratio (PSNR) are the most common methods for measuring the quality of compressed images, despite the fact that their inadequacies have long been recognized. Quality for compressed still images is sometimes evaluated using human observers who provide subjective ratings of the images. Both SNR and subjective quality judgments, however, may be inappropriate for evaluating progressive compression methods which are to be used for fast browsing applications. In this paper, we present a novel experimental and statistical framework for comparing progressive coders. The comparisons use response time studies in which human observers view a series of progressive transmissions, and respond to questions about the images as they become recognizable. We describe the framework and use it to compare several well-known algorithms (JPEG, set partitioning in hierarchical trees (SPIHT), and embedded zerotree wavelet (EZW)), and to show that a multiresolution decoding is recognized faster than a single large-scale decoding. Our experiments also show that, for the particular algorithms used, at the same PSNR, global blurriness slows down recognition more than do localized "splotch" artifacts.     

基于ROI的分形图像压缩编码

分形图像压缩由于具有非常高的压缩比越来越受人们的关注，但是高压缩比的图像方块效应非常明显，这在很大程度上影响了解码图像的质量。为了解决这个问题，借鉴JPEG2000中提出的ROI概念，提出了将感兴趣区域（ROI）图像编码与分形图像压缩编码相结合的图像编码方法，使得重构的图像中感兴趣区域的保真度高于背景区。该方法很好地解决了图像的压缩比和重构图像质量之间的矛盾。实验结果证明：此方法在获得较高压缩比的同时，压缩编码时间大为降低，解码图像质量也有较大的改善，且总体编码性能优于JPEG编码。     

Semi-supervised dimension reduction based on hypergraph embedding for hyperspectral images

Dimension reduction (DR) is an efficient and effective preprocessing step of hyperspectral images (HSIs) classification. Graph embedding is a frequently used model for DR, which preserves some geometric or statistical properties of original data set. The embedding using simple graph only considers the relationship between two data points, while in real-world application, the complex relationship between several data points is more important. To overcome this problem, we present a linear semi-supervised DR method based on hypergraph embedding (SHGE) which is an improvement of semi-supervised graph learning (SEGL). The proposed SHGE method aims to find a projection matrix through building a semi-supervised hypergraph which can preserve the complex relationship of the data and the class discrimination for DR. Experimental results demonstrate that our method achieves better performance than some existing DR methods for HSIs classification and is time saving compared with the existed method SEGL which used simple graph.