- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Large scale image retrieval with DCNN and local geometrical constraint model
摘要： Image retrieval, which refers to browse, search and retrieve the images of the same scene or object from a large database of digital images, has attracted increasing interests in recent years. This paper proposes a coarse-to-fine method for fast indexing with Deep Convolutional Neural Network(DCNN) and Local Geometrical Constraint Model. We first use a vector quantized DCNN feature descriptors and exploit enhanced Locality-sensitive hashing(LSH) techniques for fast coarse-grained retrieval. Then, we focus on obtaining high-precision preserved matches for fine-grained retrieval. This is formulated as a maximum likelihood estimation of a Bayesian model with latent variables indicating whether matches in the putative set are inliers or outliers. We impose the non-parametric global geometrical constraints on the correspondence using Tikhonov regularizers in a reproducing kernel Hilbert space. To ensure the well-posedness of the problem, we develop a local geometrical constraint that can preserve local structures among neighboring feature points, and it is also robust to a large number of outliers. The problem is solved by using the Expectation Maximization algorithm. Extensive experiments on real near-duplicate images for both feature matching and image retrieval demonstrate that the results of the proposed method outperform current state-of-the-art methods.
关键词： Image retrieval,Coarse-to-fine,Local geometrical constraint model,DCNN
更新于2025-09-23 15:23:52
-
[IEEE 2018 IEEE 13th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP) - Aristi Village, Zagorochoria, Greece (2018.6.10-2018.6.12)] 2018 IEEE 13th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP) - MindCamera: Interactive Image Retrieval and Synthesis
摘要： Composing a realistic picture according to the mind is tough work for most people. It is not only a complex operation but also a creation process from nonexistence to existence. Therefore, the core of this problem is to provide rich existing materials for stitching. We present an interactive sketch-based image retrieval and synthesis system, MindCamera. Compared with existing methods, it can use images of daily scenes as the dataset and proposes a sketch-based scene image retrieval model. Furthermore, MindCamera can blend the target object in the gradient domain to avoid the visible seam, and it introduces alpha matting to realize real-time foreground object extraction and composition. Experiments verify that our retrieval model has higher precision and provides more reasonable and richer materials for users. The practical usage demonstrates that MindCamera allows the interactive creation of complex images, and its final compositing results are natural and realistic.
关键词： image fusion,image retrieval,image segmentation
更新于2025-09-23 15:22:29
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Circular Relevance Feedback for Remote Sensing Image Retrieval
摘要： Relevance feedback (RF) is a popular reranking technique, which aims at improving the performance of image retrieval by taking the user's opinions into account. In this paper, we introduce a new RF method, named circular relevance feedback (CRF), to enhance the behavior of remote sensing image retrieval (RSIR). Instead of the manual selection used in the common RF method, we adopt the active learning (AL) algorithm to select the samples from the initial results automatically in each RF iteration. Moreover, to ensure the selected images are representative and informative enough, we choose different AL algorithms to complete the different RF processes. Finally, the contributions of all AL-driven RF methods are integrated using a circular fusion scheme. The encouraging experimental results on the ground truth RS image archive illustrate that our CRF is useful for enhancing the performance of RSIR. In addition, compared with many existing RF methods, our CRF achieves improved behavior.
关键词： Relevance feedback,remote sensing image retrieval
更新于2025-09-23 15:22:29
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Deep Semantic Hashing Retrieval of Remotec Sensing Images
摘要： Due to the rapid evolution of satellite systems, traditional nearest neighbor image retrieval methods used in large-scale image retrieval usually cause "curse of dimensionality" that leads to boosting feature storage and slow retrieval speed. The hashing method, which aims at mapping the high-dimensional data to compact binary hash codes in Hamming space and quickly calculates the Hamming distance by bit operation and XOR operation, can effectively achieve search and retrieval with remaining similarity for big data. In this paper, we propose a novel image retrieval method based on deep hashing learning, called deep semantic hashing(DSH), attempting to mining the semantic information of remote sensing(RS) images. Experiments carried out on an archive of RS images point out that DSH outperforms other methods to achieve the state-of-the-art performance in image retrieval applications.
关键词： image retrieval,semantic mining,Remote sensing,deep learning,hashing methods
更新于2025-09-23 15:21:21
-
Deep linear discriminant analysis hashing for image retrieval
摘要： Currently, due to the exponential growth of online images, it is necessary to consider image retrieval among large number of images, which is very time-consuming and unscalable. Although many hashing methods has been proposed, they did not show excellent performance in decreasing semantic loss during the process of hashing. In this paper, we propose a novel Deep Linear Discriminant Analysis Hashing(DLDAH) algorithm, which consists of Hash label generation stage and Deep hash model construction stage. In hash label generation stage, using extract image features, we construct an objective function based on Linear Discriminant Analysis(LDA), and minimize it to map image features into hash labels. In deep hash model construction stage, we use the generated hash labels to train a simple deep learning network for image hashing and get discriminative hash codes corresponding to training images. Then the deep hash model is used to map a new image feature into hash code for fast image retrieval. The scheme obtain a deep hash model which obtains deep semantic information without using network with a lot of layers, simplifying the process of mapping new images into hash codes. Experimental results show that our approach significantly outperforms state-of-art methods.
关键词： Image fingerprinting,Content based image retrieval,Deep network,Hashing
更新于2025-09-23 15:21:01
-
[IEEE 2019 IEEE International Conference on Manipulation, Manufacturing and Measurement on the Nanoscale (3M-NANO) - Zhenjiang, China (2019.8.4-2019.8.8)] 2019 IEEE International Conference on Manipulation, Manufacturing and Measurement on the Nanoscale (3M-NANO) - A Novel Architecture with Low Laser Power Based on Optical Networks-on-Chip
摘要： With the applications heterogeneous of Internet of Things (IoT) technology, the heterogeneous IoT systems generate a large number of heterogeneous datas, including videos and images. How to efficiently represent these images is an important and challenging task. As a local descriptor, the texton analysis has attracted wide attentions in the field of image processing. A variety of texton-based methods have been proposed in the past few years, which have achieved excellent performance. But, there still exists some problems to be solved, especially, it is difficult to describe the images with complex scenes from IoT. To address this problem, this paper proposes a multi-feature representation method called diagonal structure descriptor. It is more suitable for intermediate feature extraction and conducive to multi-feature fusion. Based on visual attention mechanism, five kinds of diagonal structure textons are defined by the color differences of diagonal pixels. Then, four types of visual features are extracted from the mapping sub-graphs and integrated into 1-D vector. Various experiments on three Corel-datasets demonstrate that the proposed method performs better than several state-of-the-art methods.
关键词： feature extraction,image representation,Internet of Things,image retrieval,local descriptor
更新于2025-09-23 15:19:57
-
PLT-based spectral features for texture image retrieval
摘要： Effective texture feature is an essential component in any content-based image retrieval system. In this study, new texture features based on image enhancement technique are presented. The authors have effectively exploited power-law transform (PLT) to extract new spectral texture features called PLT-based spectral features. Extensive experiments on the Brodatz texture database and Salzburg Textures image database prove the effectiveness of the proposed techniques and show that the proposed features significantly outperform the widely used Gabor and curvelet features. The proposed features are also compared with recently published Gaussian copula models of Gabor feature and local tetra patterns (LTrP). The experimental results confirm that the proposed features have more tolerance to scale, orientation and illumination distortion than the state-of-the-art Gabor, curvelet, Gaussian copula models of Gabor and LTrPs.
关键词： content-based image retrieval,curvelet features,power-law transform,Gabor features,texture feature
更新于2025-09-23 15:19:57
-
[Studies in Computational Intelligence] Computer Vision Methods for Fast Image Classi?cation and Retrieval Volume 821 ||
摘要： Computer vision and image retrieval and classification are a vital set of methods used in various engineering, scientific and business applications. In order to describe an image, visual features must be detected and described. Usually, the description is in the form of vectors. The book presents methods for accelerating image retrieval and classification in large datasets. Some of the methods (Chap. 5) are designed to work directly in relational database management systems.
关键词： computer vision,feature detection,image classification,image retrieval,relational databases
更新于2025-09-19 17:15:36
-
Fuzzy-NN approach with statistical features for description and classification of efficient image retrieval
摘要： Image retrieval based on content not only relies heavily upon the type of descriptors, but on the steps taken further. This has been an extensively utilized methodology for finding and fetching out images from the big database of images. Nowadays, a number of methodologies have been organized to increase the CBIR performance. This has an ability to recover pictures relying upon their graphical information. In the proposed method, Neuro-Fuzzy classifier and Deep Neural Network classifier are used to classify the pictures from a given dataset. The proposed approach obtained the highest accuracy in terms of Precision, Recall, and F-measure. To show the efficiency and effectiveness of proposed approach, statistical testing is used in terms of standard deviation, skewness, and kurtosis. The results reveal that the proposed algorithm outperforms other approaches using low computational efforts.
关键词： classifier.,Image retrieval,fuzzy,Deep Neural Network
更新于2025-09-19 17:15:36
-
[IEEE 2018 IEEE Conference on Virtual Reality and 3D User Interfaces (VR) - Tuebingen/Reutlingen, Germany (2018.3.18-2018.3.22)] 2018 IEEE Conference on Virtual Reality and 3D User Interfaces (VR) - AR in a Large Area Through Instance Recognition with Hybrid Sensors
摘要： This paper presents the concept of instant recognition with access points and vision for realizing time in a large indoor environment. The proposed method consists of constructing a database from data of various viewpoints and angles while taking images of the same object, and instant recognition through vision by comparing the input image with the database. The proposed method was evaluated through experiments, and the results show that recognition accuracy and computational time are less than our previous work using only vision.
关键词： INS,Affinity Propagation,Access Point,Image Retrieval
更新于2025-09-19 17:15:36