To detect defects in quartz rods using machine vision technology, we need to get the image through the camera and transmitted it to a computer for next processing. The energy of laser will be produced when it passed t...
详细信息
ISBN:
(纸本)9781538611074
To detect defects in quartz rods using machine vision technology, we need to get the image through the camera and transmitted it to a computer for next processing. The energy of laser will be produced when it passed the object, which makes it difficult to identify the resulting quartz rod boundary. This paper proposes a method of boundary detection based on Hough transform circle, and improves the Hough transform of some algorithms for solving the problem of the large amount of calculation.
We review the recent progress on the application of imageprocessing techniques to optical communication systems. The focus is placed mainly on the implementation complexity and performance of the techniques for optic...
ISBN:
(纸本)9781538679371;9781538679364
We review the recent progress on the application of imageprocessing techniques to optical communication systems. The focus is placed mainly on the implementation complexity and performance of the techniques for optical performance monitoring and the compensation of common phase error. We also briefly introduce several applications where machine learning algorithms could be beneficial to fiber-optic transmission system.
This paper proposes a new approach for single frame image super resolution using multiple ANFIS (Adaptive Network-based Fuzzy Inference System) mappings. It presents an implemented learning system that captures the re...
详细信息
ISBN:
(纸本)9781509060344
This paper proposes a new approach for single frame image super resolution using multiple ANFIS (Adaptive Network-based Fuzzy Inference System) mappings. It presents an implemented learning system that captures the relationship between a low resolution (LR) image patch space and a high resolution (HR) one given an external image database. In particular, a collected large number of LR and HR image patch pairs are divided into different groups with a clustering method. For each clustered group of the training samples, an ANFIS mapping is learned for super resolution (SR). The non-local means filter is subsequently employed to suppress the displeasing artefacts of the resulting reconstructed HR image. The proposed approach is evaluated on a range of natural images and compared with a number of existing state-of-the-art SR algorithms, demonstrating its effectiveness.
This paper proposes the use of perceptual hashing to improve minutiae extraction of fingerprint images that tolerates context-preserving operations. The hash extraction is performed after wavelet transform and singula...
详细信息
ISBN:
(纸本)9781509045594
This paper proposes the use of perceptual hashing to improve minutiae extraction of fingerprint images that tolerates context-preserving operations. The hash extraction is performed after wavelet transform and singular value decomposition (SVD). The performance evaluation of this approach has been assessed using various metrics such as SSIM and PSNR. Experimentally, it has shown robustness against imageprocessing operations and geometric attacks.
High resolution image handling often results with high energy burden for battery-powered devices, such as sensor nodes in WSN. Motivation for this study is assessment of energy consumption of the sensor node with high...
详细信息
ISBN:
(纸本)9783319615639;9783319615622
High resolution image handling often results with high energy burden for battery-powered devices, such as sensor nodes in WSN. Motivation for this study is assessment of energy consumption of the sensor node with high-resolution camera, featuring imageprocessing. We present a selection of object detection algorithms and evaluate their efficiency. To verify applicability of those algorithms, we acquired image sequence that correspond to applications of pests detection in agriculture. We verified considered algorithms' performances: recall, precision and expected reduction of the data amount. Energy required to execute considered algorithms was measured on ARM processor based platform. Our results show that object extraction on a node can provide reduction of the data amount by up to three orders of magnitude. While simple algorithms can lead to lower overall energy consumption of the node, the more complex algorithm provides better performances, but at a cost of prohibitively high energy consumption.
Traditional dehazing techniques, as a well-studied topic in imageprocessing, are now widely used to eliminate the haze effects from individual images. However, even the state-of-the-art dehazing algorithms may not pr...
详细信息
ISBN:
(纸本)9781538632215
Traditional dehazing techniques, as a well-studied topic in imageprocessing, are now widely used to eliminate the haze effects from individual images. However, even the state-of-the-art dehazing algorithms may not provide sufficient support to video analytics, as a crucial pre-processing step for video-based decision making systems (e.g., robot navigation), due to the limitations of these algorithms on poor result coherence and low processing efficiency. This paper presents a new framework, particularly designed for video dehazing, to output coherent results in real time, with two novel techniques. We decompose the dehazing algorithms into three generic components, namely transmission map estimator, atmospheric light estimator and haze-free image generator. They can be simultaneously processed by multiple threads in the distributed system, such that the processing efficiency is optimized by automatic CPU resource allocation based on the workloads. The combination of these techniques enables our framework to generate highly consistent and accurate dehazing results in real-time, by using only 3 PCs connected by Ethernet.
The paper discusses the application of System on Chip devices for processing Megapixel video streams. The domain of imageprocessing using high resolution images is very demanding in the scope of calculating power and...
详细信息
ISBN:
(纸本)9781538624029
The paper discusses the application of System on Chip devices for processing Megapixel video streams. The domain of imageprocessing using high resolution images is very demanding in the scope of calculating power and frequently exploits special processing hardware. The progress of integration technology brings about SoC which are capable of meeting such processing demands. Characteristics of FPGA and DSP based systems are assessed. A concise survey of chip solutions is presented and a pair of representative systems is chosen for comparison of their properties. The design of a smoke detection system is used for evaluating the advantages and drawbacks of using such SoC solutions. Almost identical imageprocessingalgorithms are implemented in the case of FPGA and DSP. Both solutions meet the imageprocessing requirements and are suitable for real time processing. FPGA based solution is much faster as it uses pipelining for improving the utilisation of scarce memory resources.
In image tamper detection and recovery methods prevalently two sets of data need to be embedded into the main image, one known as authentication data for tamper detection and the other known as reference data for tamp...
详细信息
ISBN:
(纸本)9781538649725
In image tamper detection and recovery methods prevalently two sets of data need to be embedded into the main image, one known as authentication data for tamper detection and the other known as reference data for tamper recovery. In this paper, a scalable method for image tamper detection and recovery based on a dual-rate source-channel coding is proposed. In the proposed method the original image is first compressed by a scalable source coding algorithm. The compressed bitstream is then partitioned into two parts. Both parts are protected with a channel coding algorithm but with two different rates according to their importance for tamper recovery process. The proposed method takes advantage of two state-of-the-art algorithms, SPIRT for source coding and LDPC for channel coding. Simulation results show a noticeable improvement compared with related tamper detection and recovery schemes in the literature. Besides the reconstruction quality of the recovered image is scalable. This means that in low tampering rates, high-quality images are recovered and in high tampering rates, the image is still recoverable but with less reconstruction quality. Therefore the proposed method succeeded to both recover higher tampering rates and preserve the image quality.
Traffic Light detection and recognition has become one of the critical research areas when it comes to autonomous driving vehicles. Developing an algorithm to detect different shaped traffic lights such as traffic lig...
详细信息
ISBN:
(纸本)9781538673539;9781538673522
Traffic Light detection and recognition has become one of the critical research areas when it comes to autonomous driving vehicles. Developing an algorithm to detect different shaped traffic lights such as traffic lights (TLs) with arrowheads, circular shapes, horizontally and vertically oriented TLs and TLs in different illumination conditions during daytime still remains a challenge. This paper evaluates the existing algorithms, which uses heuristic or template matching based imageprocessing methods and other learning based systems, that consumes higher time and processing power for training and map based systems and proposes a novel optimized algorithm using machine vision techniques to meet those challenges. The experiments were carried out using videos taken by front camera of vehicles. The results of the algorithm show a higher accuracy for precision and recall evaluations.
In the field of approximate nearest neighbor (ANN) search, rare of the existing approaches are tailored for video applications. The Ring Intersection Approximate Nearest Neighbor (RIANN) is the first ANN search algori...
详细信息
In the field of approximate nearest neighbor (ANN) search, rare of the existing approaches are tailored for video applications. The Ring Intersection Approximate Nearest Neighbor (RIANN) is the first ANN search algorithm for videos. It achieves real-time by performing the ANN search on the sparse grid and interpolating others. For some applications, the dense ANN search is needed to ensure the searching accuracy. To achieve dense ANN search in real-time, we consider the parallel computing as a solution. However, the RIANN algorithm is not suitable for parallel computing as the algorithm itself suffers from bad thread coherency. In this paper, we propose the Sphere Ring Intersection Approximate Nearest Neighbor (SRIANN), which solves the problem of bad thread coherency and improves the accuracy of ANN search compared to the original RIANN method. The experimental results show that the proposed method is the only one able to perform dense ANN search for CIF videos in real-time.
暂无评论