Structural image retrieval using automatic image annotation and region based inverted file

Image retrieval has lagged far behind text retrieval despite more than two decades of intensive research effort. Most of the research on image retrieval in the last two decades are on content based image retrieval or image retrieval based on low level features. Recent research in this area focuses on semantic image retrieval using automatic image annotation. Most semantic image retrieval techniques in literature, however, treat an image as a bag of features/words while ignore the structural or spatial information in the image. In this paper, we propose a structural image retrieval method based on automatic image annotation and region based inverted file. In the proposed system, regions in an image are treated the same way as keywords in a structural text document, semantic concepts are learnt from image data to label image regions as keywords and weight is assigned to each keyword according to spatial position and relationship. As the result, images are indexed and retrieved in the same way as structural document retrieval. Specifically, images are broken down to regions which are represented using colour, texture and shape features. Region features are then quantized to create visual dictionaries which are similar to monolingual dictionaries like English or Chinese dictionaries. In the next step, a semantic dictionary similar to a bilingual dictionary like the English–Chinese dictionary is learnt to mapping image regions to semantic concepts. Finally, images are then indexed and retrieved using a novel region based inverted file data structure. Results show the proposed method has significant advantage over the widely used Bayesian annotation models.     

基于非对称监督深度离散哈希的图像检索

哈希广泛应用于图像检索任务。针对现有深度监督哈希方法的局限性,该文提出了一种新的非对称监督深度离散哈希(ASDDH)方法来保持不同类别之间的语义结构,同时生成二进制码。首先利用深度网络提取图像特征,根据图像的语义标签来揭示每对图像之间的相似性。为了增强二进制码之间的相似性,并保证多标签语义保持,该文设计了一种非对称哈希方法,并利用多标签二进制码映射,使哈希码具有多标签语义信息。此外,引入二进制码的位平衡性对每个位进行平衡,鼓励所有训练样本中的–1和+1的数目近似。在两个常用数据集上的实验结果表明,该方法在图像检索方面的性能优于其他方法。     

Learning semantic and visual similarity for endomicroscopy video retrieval

Content-based image retrieval (CBIR) is a valuable computer vision technique which is increasingly being applied in the medical community for diagnosis support. However, traditional CBIR systems only deliver visual outputs, i.e., images having a similar appearance to the query, which is not directly interpretable by the physicians. Our objective is to provide a system for endomicroscopy video retrieval which delivers both visual and semantic outputs that are consistent with each other. In a previous study, we developed an adapted bag-of-visual-words method for endomicroscopy retrieval, called "Dense-Sift," that computes a visual signature for each video. In this paper, we present a novel approach to complement visual similarity learning with semantic knowledge extraction, in the field of in vivo endomicroscopy. We first leverage a semantic ground truth based on eight binary concepts, in order to transform these visual signatures into semantic signatures that reflect how much the presence of each semantic concept is expressed by the visual words describing the videos. Using cross-validation, we demonstrate that, in terms of semantic detection, our intuitive Fisher-based method transforming visual-word histograms into semantic estimations outperforms support vector machine (SVM) methods with statistical significance. In a second step, we propose to improve retrieval relevance by learning an adjusted similarity distance from a perceived similarity ground truth. As a result, our distance learning method allows to statistically improve the correlation with the perceived similarity. We also demonstrate that, in terms of perceived similarity, the recall performance of the semantic signatures is close to that of visual signatures and significantly better than those of several state-of-the-art CBIR methods. The semantic signatures are thus able to communicate high-level medical knowledge while being consistent with the low-level visual signatures and much shorter than them. In our resulting retrieval system, we decide to use visual signatures for perceived similarity learning and retrieval, and semantic signatures for the output of an additional information, expressed in the endoscopist own language, which provides a relevant semantic translation of the visual retrieval outputs.     

A CBIR system based on saliency driven local image features and multi orientation texture features

In Content-based Image Retrieval (CBIR), the user provides the query image in which only a selective portion of the image carries the foremost vital information known as the object region of the image. However, the human visual system also focuses on a particular salient region of an image to instinctively understand its semantic meaning. Therefore, the human visual attention technique can be well imposed in the CBIR scheme. Inspired by these facts, we initially utilized the signature saliency map-based approach to decompose the image into its respective main object region (ObR) and non-object region (NObR). ObR possesses most of the vital image information, so block-level normalized singular value decomposition (SVD) has been used to extract salient features of the ObR. In most natural images, NObR plays a significant role in understanding the actual semantic meaning of the image. Accordingly, multi-directional texture features have been extracted from NObR using Gabor filter on different wavelengths. Since the importance of ObR and NObR features are not equal, a new homogeneity-based similarity matching approach has been devised to enhance retrieval accuracy. Finally, we have demonstrated retrieval performances using both the combined and distinct ObR and NObR features on seven standard coral, texture, object, and heterogeneous datasets. The experimental outcomes show that the proposed CBIR system has a promising retrieval efficiency and outperforms various existing systems substantially.     

Image annotation using high order statistics in non-Euclidean spaces

Automatic image annotation is a promising way to achieve more effective image retrieval and image analysis by using keywords associated to the image content. Due to the semantic gap between low-level visual features and high-level semantic concepts of an image, however, the performances of many existing algorithms are not so satisfactory. In this paper, a novel image classification scheme, named high order statistics based maximum a posterior (HOS-MAP), is proposed to deal with the issue of image annotation. To bridge the gap between human judgment and machine intelligence, the proposed scheme first constructs a dissimilarity representation for each image in a non-Euclidean space; then, the information of dissimilarity diffusion distribution for each image is achieved with respect to the high-order statistics of a triplet of nearest neighbor images; finally, a maximum a posteriori algorithm with the information of Gaussian Mixture Model and dissimilarity diffusion distribution is adopted to estimate the relevance between each annotation and an input un-annotated image. Experimental results on a general-purpose image database demonstrate the effectiveness and efficiency of the proposed automatic image annotation scheme.     

Discrete hashing with triple supervision learning

In recent years, discrete supervised hashing methods have attracted increasing attention because of their high retrieval efficiency and precision. However, in these methods, some effective semantic information is typically neglected, which means that all the information is not sufficiently utilized. Moreover, these methods often only decompose the first-order features of the original data, ignoring the more fine-grained higher-order features. To address these problems, we propose a supervised hashing learning method called discrete hashing with triple supervision learning (DHTSL). Specifically, we integrate three aspects of semantic information into this method: (1) the bidirectional mapping of semantic labels; (2) pairwise similarity relations; (3) second-order features from the original data. We also design a discrete optimization method to solve the proposed objective function. Moreover, an out-of-sample extension strategy that can better maintain the independence and balance of hash codes is employed to improve retrieval performance. Extensive experiments on three widely used datasets demonstrate its superior performance.     

基于深度哈希算法的极光图像分类与检索方法

面对形态万千、变化复杂的海量极光数据,对其进行分类与检索为进一步研究地球磁场物理机制和空间信息具有重要意义。该文基于卷积神经网络(CNN)对图像特征提取方面的良好表现,以及哈希编码可以满足大规模图像检索对检索时间的要求,提出一种端到端的深度哈希算法用于极光图像分类与检索。首先在CNN中嵌入空间金字塔池化(SPP)和幂均值变换(PMT)来提取图像中多种尺度的区域信息;其次在全连接层之间加入哈希层,将全连接层最能表现图像的高维语义信息映射为紧凑的二值哈希码,并在低维空间使用汉明距离对图像对之间的相似性进行度量;最后引入多任务学习机制,充分利用图像标签信息和图像对之间的相似度信息来设计损失函数,联合分类层和哈希层的损失作为优化目标,使哈希码之间可以保持更好的语义相似性,有效提升了检索性能。在极光数据集和 CIFAR-10 数据集上的实验结果表明,所提出方法检索性能优于其他现有检索方法,同时能够有效用于极光图像分类。     

Deep multi query image retrieval

There exist few studies investigating the multi-query image retrieval problem. Existing methods are not based on hash codes. As a result, they are not efficient and fast. In this study, we develop an efficient and fast multi-query image retrieval method when the queries are related to more than one semantic. Image hash codes are generated by a deep hashing method. Consequently, the method requires lower storage space, and it is faster compared to the existing methods. The retrieval is based on the Pareto front method. Reranking performed on the retrieved images by using non-binary deep-convolutional features increase retrieval accuracy considerably. Unlike previous studies, the method supports an arbitrary number of queries. It outperforms similar multi-query image retrieval studies in terms of retrieval time and retrieval accuracy.     

A Scheme for Intelligent Image Retrieval in Multimedia Databases

Storage and retrieval of visual data play an important role in multimedia systems. We have developed a content-based scheme for retrieving images from multimedia databases intelligently. The retrieval takes two stages. The first stage retrieves an image based on partial information. In the second stage, the system accumulates knowledge from the results of the first-stage retrieval. It analyzes the subspace of features from the resulting images and tries to understand the query request. It also makes full use of the entire index space, although queries can be made on partial information. The technology developed will find many applications in multimedia areas. It will also provide a tool for studying how humans rank the similarity of images and what information people use in visual perception, etc., and will help in the development of methods based on these human approaches.     

A framework for medical image retrieval using machine learning and statistical similarity matching techniques with relevance feedback.

A content-based image retrieval (CBIR) framework for diverse collection of medical images of different imaging modalities, anatomic regions with different orientations and biological systems is proposed. Organization of images in such a database (DB) is well defined with predefined semantic categories; hence, it can be useful for category-specific searching. The proposed framework consists of machine learning methods for image prefiltering, similarity matching using statistical distance measures, and a relevance feedback (RF) scheme. To narrow down the semantic gap and increase the retrieval efficiency, we investigate both supervised and unsupervised learning techniques to associate low-level global image features (e.g., color, texture, and edge) in the projected PCA-based eigenspace with their high-level semantic and visual categories. Specially, we explore the use of a probabilistic multiclass support vector machine (SVM) and fuzzy c-mean (FCM) clustering for categorization and prefiltering of images to reduce the search space. A category-specific statistical similarity matching is proposed in a finer level on the prefiltered images. To incorporate a better perception subjectivity, an RF mechanism is also added to update the query parameters dynamically and adjust the proposed matching functions. Experiments are based on a ground-truth DB consisting of 5000 diverse medical images of 20 predefined categories. Analysis of results based on cross-validation (CV) accuracy and precision-recall for image categorization and retrieval is reported. It demonstrates the improvement, effectiveness, and efficiency achieved by the proposed framework.     

A novel unsupervised multiple feature hashing for image retrieval and indexing (MFHIRI)

Recently, techniques that can automatically figure out the incisive information from gigantic visual databases are urging popularity. The existing multi-feature hashing method has achieved good results by fusing multiple features, but in processing these multi-features, fusing multi-features into one feature will cause the feature dimension to be very high, increasing the amount of calculation. On the one hand, it is not easy to discover the internal ties between different features. This paper proposes a novel unsupervised multiple feature hashing for image retrieval and indexing (MFHIRI) method to learn multiple views in a composite manner. The proposed scheme learns the binary codes of various information sources in a composite manner, and our scheme relies on weighted multiple information sources and improved KNN concept. In particular, here we adopt an adaptive weighing scheme to preserve the similarity and consistency among binary codes. Precisely, we follow the graph modeling theory to construct improved KNN concept, which further helps preserve different statistical properties of individual sources. The important aspect of improved KNN scheme is that we can find the neighbors of a data point by searching its neighbors’ neighbors. During optimization, the sub-problems are solved in parallel which efficiently lowers down the computation cost. The proposed approach shows consistent performance over state-of-the-art (three single-view and eight multi-view approaches) on three broadly followed datasets viz. CIFAR-10, NUS-WIDE and Caltech-256.     

A unified image retrieval framework on local visual and semantic concept-based feature spaces

This paper presents a learning-based unified image retrieval framework to represent images in local visual and semantic concept-based feature spaces. In this framework, a visual concept vocabulary (codebook) is automatically constructed by utilizing self-organizing map (SOM) and statistical models are built for local semantic concepts using probabilistic multi-class support vector machine (SVM). Based on these constructions, the images are represented in correlation and spatial relationship-enhanced concept feature spaces by exploiting the topology preserving local neighborhood structure of the codebook, local concept correlation statistics, and spatial relationships in individual encoded images. Finally, the features are unified by a dynamically weighted linear combination of similarity matching scheme based on the relevance feedback information. The feature weights are calculated by considering both the precision and the rank order information of the top retrieved relevant images of each representation, which adapts itself to individual searches to produce effective results. The experimental results on a photographic database of natural scenes and a bio-medical database of different imaging modalities and body parts demonstrate the effectiveness of the proposed framework.     

图像语义自描述性的实现方法研究

根据语义特征进行检索是多媒体检索技术的发展趋势.图像是最重要的媒体之一,本文提出了一种实现图像语义自描述性的方法.通过在JPEG文件中嵌入描述图像语义的XML文件,使计算机可以对图像进行语义级的检索.实验表明,该方法创建的语义描述可以从图像文件中读取,具有标准化、结构性和可读性的特点,尤其适用于在Internet上检索图像.该方法对于提高图像信息的利用率具有重要的意义.     

Unifying keywords and visual contents in image retrieval

We're interested in using keywords and visual content together in image retrieval. We used a seamless joint querying and relevance feedback scheme based on keywords and lowlevel visual content, incorporating keyword similarities. We developed an algorithm for a learned word similarity matrix and conducted experiments that validated our approach     

Using binarization and hashing for efficient SIFT matching

The well-known SIFT is capable of extracting distinctive features for image retrieval. However, its matching is time consuming and slows down the entire process. In the SIFT matching, the Euclidean distance is used to measure the similarity of two features, which is expensive because it involves taking square root. Moreover, the scale of the image database is usually too large to adopt linear search for image retrieval. To improve the SIFT matching, this paper proposes a fast image retrieval scheme transforms the SIFT features to binary representations. The complexity of the distance calculation is reduced to bit-wise operation and the retrieval time is greatly decreased. Moreover, the proposed scheme utilizes hashing for retrieving similar images according to the binarized features and further speeds up the retrieval process. The experiment results show the proposed scheme can retrieve images efficiently with only a little sacrifice of accuracy as compared to SIFT.     

基于改进哈希算法的图像检索方法

当前主流图像检索技术所采用的传统视觉特征编码缺少足够的学习能力,影响学习得到的特征表达能力。此外,由于视觉特征维数高,会消耗大量的内存,因此降低了图像检索的性能。文中基于深度卷积神经网络与改进的哈希算法,提出并设计了一种端到端训练方式的图像检索方法。该方法将卷积神经网络提取的高层特征和哈希函数相结合,学习到具有足够表达能力的哈希特征,从而在低维汉明空间中完成对图像数据的大规模检索。在两个常用数据集上的实验结果表明,所提出的哈希图像检索方法的检索性能优于当前的一些主流方法。     

Knowledge representation and sharing using visual semantic modeling for diagnostic medical image databases.

Information technology offers great opportunities for supporting radiologists' expertise in decision support and training. However, this task is challenging due to difficulties in articulating and modeling visual patterns of abnormalities in a computational way. To address these issues, well established approaches to content management and image retrieval have been studied and applied to assist physicians in diagnoses. Unfortunately, most of the studies lack the flexibility of sharing both explicit and tacit knowledge involved in the decision making process, while adapting to each individual's opinion. In this paper, we propose a knowledge repository and exchange framework for diagnostic image databases called "evolutionary system for semantic exchange of information in collaborative environments" (Essence). This framework uses semantic methods to describe visual abnormalities, and offers a solution for tacit knowledge elicitation and exchange in the medical domain. Also, our approach provides a computational and visual mechanism for associating synonymous semantics of visual abnormalities. We conducted several experiments to demonstrate the system's capability of matching synonym terms, and the benefit of using tacit knowledge in improving the meaningfulness of semantic queries.     

Effective image retrieval based on hidden concept discovery in image database.

This paper addresses content-based image retrieval in general, and in particular, focuses on developing a hidden semantic concept discovery methodology to address effective semantics-intensive image retrieval. In our approach, each image in the database is segmented into regions associated with homogenous color, texture, and shape features. By exploiting regional statistical information in each image and employing a vector quantization method, a uniform and sparse region-based representation is achieved. With this representation, a probabilistic model based on statistical-hidden-class assumptions of the image database is obtained, to which the expectation-maximization technique is applied to analyze semantic concepts hidden in the database. An elaborated retrieval algorithm is designed to support the probabilistic model. The semantic similarity is measured through integrating the posterior probabilities of the transformed query image, as well as a constructed negative example, to the discovered semantic concepts. The proposed approach has a solid statistical foundation; the experimental evaluations on a database of 10000 general-purposed images demonstrate its promise and effectiveness.