- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2019 IEEE 8th International Conference on Advanced Optoelectronics and Lasers (CAOL) - Sozopol, Bulgaria (2019.9.6-2019.9.8)] 2019 IEEE 8th International Conference on Advanced Optoelectronics and Lasers (CAOL) - Improving the accuracy of the laser control system
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,local descriptor,image representation,image retrieval,Internet of Things
更新于2025-09-19 17:13:59
-
[Lecture Notes in Electrical Engineering] Engineering Vibration, Communication and Information Processing Volume 478 (ICoEVCI 2018, India) || Color Histogram- and Smartphone-Based Diabetic Retinopathy Detection System
摘要： Diabetic Retinopathy (DR) is diabetes-related eye disorder. An initial eye examination is the ?nest method to avoid DR. In this paper, a low-price DR detection algorithm using mobile phone-and color histogram-based image retrieval technique has been proposed. The mobile phone will take a picture of the patient’s eye with the help of 20D condensing lens, and then implement a color histogram retrieval program to ?nd the similar picture from the collected database. The presented system reduces the professional’s work of DR identi?cation. Our aim is to make an effective, easy, and low-cost eye examination program, which is ideal for underdeveloped regions and make it available to one and all.
关键词： Histogram,Image retrieval,Diabetic retinopathy,Smartphone
更新于2025-09-10 09:29:36
-
An innovative method of retrieving images through clusters, means and wavelet transformation
摘要： This paper introduces a new method CLMWT(cluster local mean wavelet transform) using the primitive features like color, texture and shape in which the features are extracted by using different components of an image using various methods clustering, local mean histogram and wavelet transform. This manuscript exhibit a technique CLMWT to extort texture, color and shape features of an image hastily for content based image retrieval. First clustering is done for the image and then local mean is applied and based on wavelet transform technique compression is done and the mean is calculated for the compressed image. Related to this way of extraction a CBIR method is intended with color, texture and shape by forming the mean of the feature vector. The proposed work CLMWT checks its performance of the method with other methods accordingly this approach gives better performance than using two combinations.
关键词： Histogram,Content based,Image retrieval,Color,Texture,Shape,Wavelet,Local mean
更新于2025-09-10 09:29:36
-
La recuperación automatizada de imágenes: retos y soluciones
摘要： Analysis of the peculiar characteristics of images as documents and its consequences for the development of automated retrieval of this type of documentation. After developing the concepts of image and image retrieval, the challenges that image presents from the point of view of retrieval are analyzed, especially the semantic gap, and the main solutions found in the literature since 1990 to present are described. The current approach (SBVIR) is characterized by the simultaneous employment of the visual code and the language in representing images.
关键词： Content-Based Image Retrieval,Image Retrieval,Photography,Information Retrieval,Image
更新于2025-09-09 09:28:46
-
Exporting Diabetic Retinopathy Images from VA VistA Imaging for Research
摘要： The US Department of Veterans Affairs has been acquiring store and forward digital diabetic retinopathy surveillance retinal fundus images for remote reading since 2007. There are 900+ retinal cameras at 756 acquisition sites. These images are manually read remotely at 134 sites. A total of 2.1 million studies have been performed in the teleretinal imaging program. The human workload for reading images is rapidly growing. It would be ideal to develop an automated computer algorithm that detects multiple eye diseases as this would help standardize interpretations and improve efficiency of the image readers. Deep learning algorithms for detection of diabetic retinopathy in retinal fundus photographs have been developed and there are needs for additional image data to validate this work. To further this research, the Atlanta VA Health Care System (VAHCS) has extracted 112,000 DICOM diabetic retinopathy surveillance images (13,000 studies) that can be subsequently used for the validation of automated algorithms. An extensive amount of associated clinical information was added to the DICOM header of each exported image to facilitate correlation of the image with the patient’s medical condition. The clinical information was saved as a JSON object and stored in a single Unlimited Text (VR = UT) DICOM data element. This paper describes the methodology used for this project and the results of applying this methodology.
关键词： VistA,Image retrieval for research,DICOM,Retinal imaging,JSON,Diabetic retinopathy
更新于2025-09-09 09:28:46
-
[Studies in Computational Intelligence] Recent Advances in Computer Vision Volume 804 (Theories and Applications) || Content-Based Image Retrieval Using Multiresolution Feature Descriptors
摘要： The advent of low-cost cameras and smartphones have made the task of image capturing quite easy nowadays. This has resulted in the collection of large number of unorganized images. Accessing images from large repository of unorganized images is quite challenging. There is a need of such systems which help in proper organization and easy access of images. The field of image retrieval, using text or image, attempts to solve this problem. While text-based retrieval systems are quite popular, they suffer from certain drawbacks. The other type of image retrieval system, which is Content-based Image Retrieval (CBIR) system, uses image features to search for relevant images. This chapter discusses the concept multiresolution feature descriptors for CBIR. For capturing varying level of details, single resolution processing of image proves to be insufficient. The use of multiresolution descriptors prove to be quite efficient in capturing complex foreground and background details in an image. This chapter discusses the important properties and advantages of multiresolution feature descriptors. Furthermore, this chapter proposes a CBIR technique using a novel multiresolution feature descriptor. The proposed method constructs feature vector by capturing shape feature in a localized manner. The experimental results show the effectiveness of the proposed method.
关键词： Content-Based Image Retrieval,CBIR,Feature Descriptors,Multiresolution Feature Descriptors,Image Retrieval
更新于2025-09-04 15:30:14
-
[Lecture Notes in Computer Science] Pattern Recognition and Computer Vision Volume 11257 (First Chinese Conference, PRCV 2018, Guangzhou, China, November 23-26, 2018, Proceedings, Part II) || Deep Supervised Auto-encoder Hashing for Image Retrieval
摘要： Image hashing approaches map high dimensional images to compact binary codes that preserve similarities among images. Although the image label is important information for supervised image hashing methods to generate hashing codes, the retrieval performance will be limited according to the performance of the classi?er. Therefore, an e?ective supervised auto-encoder hashing method (SAEH) is proposed to generate low dimensional binary codes in a point-wise manner through deep convolutional neural network. The auto-encoder structure in SAEH is designed to simultaneously learn image features and generate hashing codes. Moreover, some extra relaxations for generating binary hash codes are added to the objective function. The extensive experiments on several large scale image datasets validate that the auto-encoder structure can indeed increase the performance for supervised hashing and SAEH can achieve the best image retrieval results among other prominent supervised hashing methods.
关键词： Image hashing,Image retrieval,Supervised learning,Deep neural network,Convolutional auto-encoder
更新于2025-09-04 15:30:14
-
[IEEE 2018 24th International Conference on Pattern Recognition (ICPR) - Beijing, China (2018.8.20-2018.8.24)] 2018 24th International Conference on Pattern Recognition (ICPR) - Deep High-order Supervised Hashing for Image Retrieval
摘要： Recently, deep hashing has achieved excellent performances in large-scale image retrieval by simultaneously learning deep features and hashing function. However, state-of-the-art works have so far failed to explore the feature statistics higher than first-order. In this paper, to take a step towards addressing this problem, we propose two novel Deep High-order Supervised Hashing architectures (DHoSH), i.e., point-wise labels based DHoSH (DHoSH-PO) and pair-wise labels based DHoSH (DHoSH-PA). The core of DHoSH is that a trainable layer of bilinear pooling incorporates into deep convolutional neural networks (CNNs) for end-to-end learning. This layer captures the local feature interactions of the image by outer product, employing the autocorrelation information and cross-correlation information of deep features. Furthermore, our DHoSH method systematically exploits the high-order statistics of features of multiple layers. Extensive experiments on commonly used benchmarks illuminate that both DHoSH-PO and DHoSH-PA can obtain competitive improvements over its first-order counterparts, and achieve state-of-the-art performance for image retrieval task.
关键词： image retrieval,supervised learning,deep hashing,high-order statistics,bilinear pooling
更新于2025-09-04 15:30:14
-
[IEEE 2018 24th International Conference on Pattern Recognition (ICPR) - Beijing, China (2018.8.20-2018.8.24)] 2018 24th International Conference on Pattern Recognition (ICPR) - Learning Cross-Modal Deep Embeddings for Multi-Object Image Retrieval using Text and Sketch
摘要： In this work we introduce a cross modal image retrieval system that allows both text and sketch as input modalities for the query. A cross-modal deep network architecture is formulated to jointly model the sketch and text input modalities as well as the the image output modality, learning a common embedding between text and images and between sketches and images. In addition, an attention model is used to selectively focus the attention on the different objects of the image, allowing for retrieval with multiple objects in the query. Experiments show that the proposed method performs the best in both single and multiple object image retrieval in standard datasets.
关键词： cross-modal,image retrieval,sketch,attention model,deep learning,text
更新于2025-09-04 15:30:14
-
Augmented reality system for facility management using image-based indoor localization
摘要： Image-based localization has provided opportunities for efficient facility management. Combined with augmented reality (AR), automated localization can offer visually assistive information in facility management. However, implementing an AR-based facility management system with image-based localization is difficult. Device-intensive methods or markers were prerequisites for facility information display. Localization accuracy and information readability and accessibility were some of the issues to be resolved for a successful representation of facility information. This paper presents an AR system for facility management using an image-based indoor localization method that estimates the user's indoor position and orientation by comparing the user's perspective to building information modeling (BIM) based on a deep learning computation. A graphics processing unit (GPU)-enabled server is used for the deep learning computation, and the resultant information is wirelessly transferred to the mobile AR device through transmission control protocol/Internet protocol (TCP/IP). Thereafter, spatial mapping visually fits the object of interest (e.g. pipes) onto the AR image using three-dimensional (3D) sensing capability of AR device. Experts evaluated that the proposed system has potential for improved facility management and identified future research direction, such as integrated information presentation and effective reflection of rehabilitation efforts on the drawings.
关键词： Facility management,Indoor localization,Image retrieval,Augmented reality
更新于2025-09-04 15:30:14