Color is a powerful feature for image analysis but it is usually not used in image classification schemes. We propose a method to combine fuzzy color information with the result obtained from a One-Versus-All classifi...
详细信息
ISBN:
(纸本)9781509060344
Color is a powerful feature for image analysis but it is usually not used in image classification schemes. We propose a method to combine fuzzy color information with the result obtained from a One-Versus-All classifier (OVA) trained with Bag-of-features. This method consists in weighting the outputs of the OVA classifier based on the distances between the new image to be classified and the classes. Experimental results show that our approach improves OVA classifier performance.
Sustainable agriculture is an important field where not much attention is given though it is highly necessary, so as to monitor the growth of crops for their efficient growth in most nutritious ways. For effective gro...
详细信息
ISBN:
(纸本)9781538658741
Sustainable agriculture is an important field where not much attention is given though it is highly necessary, so as to monitor the growth of crops for their efficient growth in most nutritious ways. For effective growth of crops, lot of chemicals like fertilizers and pesticides are used, however, excessive usage of them results in damage to land and water resources. The attack of pests is a major criteria which affect crop yield. Various crop monitoring technologies are available which are highly expensive and not all farmers can afford. Moreover in India, farmers are not capable of understanding the operation and handling of such sophisticated technology. In this paper, we propose a system which is cheaper and easy to operate with multiple application. The proposed system uses a technology which utilizes machine learning and ANN algorithms using UAV that helps us to locate regions that are affected by diseases and pesticides so that we can particularly focus on the regions that are affected and apply chemicals only in that particular area, with the entire system being cost effective. For this purpose, we divide the entire area into n x n segments and using imageprocessing, the segmented areas are analysed and processed using Ardupilot and a central operating system to monitor using python and open CV, giving the simplest user interface for monitoring purposes.
Least mean square (LMS) algorithm is very familiar and used successfully in adaptive filters which are widely used in several areas such as wireless communication or imageprocessing. However, it is inefficient for so...
详细信息
ISBN:
(纸本)9781538664698
Least mean square (LMS) algorithm is very familiar and used successfully in adaptive filters which are widely used in several areas such as wireless communication or imageprocessing. However, it is inefficient for some situations such as the constant step-size in its update equation or eigenvalue spread of the input autocorrelation matrix. Besides, the performance of the LMS algorithm deteriorates for a system identification problem when we deal with a sparse system in which the most of the coefficients are zero or near zero. In this work a new sparsity based LMS-type algorithm has been proposed. It exploits the advantages of the recently proposed q-LMS algorithm which is based on q-calculus, a variable step-size LMS algorithm in which a function is used to adjust the step-size and the zeroattraction factor derived by an additional l 0 norm in its error function. The proposed algorithm has been compared with both q-LMS algorithm and zero-attracting function controlled variable step-size LMS algorithms according to the convergence speed and mean square deviation. Experiments showed that the new algorithm has a faster convergence without sacrificing from mean square deviation level for a sparse system identification with an uncorrelated or correlated input signals.
In this paper we developed the algorithm for the automatic tracking of unmanned aircraft for image sequence based on correlation filtering to improve the efficiency of conducting objective control and simulation in Ma...
详细信息
ISBN:
(纸本)9781538605219
In this paper we developed the algorithm for the automatic tracking of unmanned aircraft for image sequence based on correlation filtering to improve the efficiency of conducting objective control and simulation in Matlab and C#. The measurement results showed good convergence of the theoretical and experimental data. The advantage of this approach is the possibility to use it in real-time through the application of fast direct and inverse Fourier transform;availability of the implementation in various environments for object-oriented programming such as Visual C#/C using the library of structures and algorithms EmguCV/OpenCV and the libraries of algorithms for fast discrete Fourier transform FFTWSharp/FFTW;the possibility of the approach implementing in the systems with reconfigurable integrated circuits based on programmable logic;"Field programmable gate array" (FPGA);the use of the adaptive approach in the program code design. For further improvement of the algorithm it is appropriate to work in the area of window tracking changes due to the deformation or removal of the object to reduce the computational cost.
The problem of searching a digital image in a very huge database is called content-based image retrieval (CBIR). Texture represents spatial or statistical repetition in pixel intensity and orientation. When abnormal c...
详细信息
ISBN:
(纸本)9789811021046;9789811021039
The problem of searching a digital image in a very huge database is called content-based image retrieval (CBIR). Texture represents spatial or statistical repetition in pixel intensity and orientation. When abnormal cells form within the brain is called brain tumor. In this paper, we have developed a texture feature extraction of MRI brain tumor image retrieval. There are two parts, namely feature extraction process and classification. First, the texture features are extracted using techniques like curvelet transform, contourlet transform, and Local Ternary Pattern (LTP). Second, the supervised learning algorithms like Deep Neural Network (DNN) and Extreme Learning Machine (ELM) are used to classify the brain tumor images. The experiment is performed on a collection of 1000 brain tumor images with different modalities and orientations. Experimental results reveal that contourlet transform technique provides better than curvelet transform and local ternary pattern.
To solve the problem that traditional geometric correction algorithms of QR code will be influenced by light, shooting impact angle, correction algorithm robustness and so on, this paper proposes an adaptive algorithm...
详细信息
ISBN:
(纸本)9781538608432
To solve the problem that traditional geometric correction algorithms of QR code will be influenced by light, shooting impact angle, correction algorithm robustness and so on, this paper proposes an adaptive algorithm based on image gray features, which achieves the accurate reading and rapid correction of QR code location information. After the pre-processing of QR code image, this algorithm calculates the dynamic threshold according to the gray value of feature image and the calculated threshold is used to confirm the most appropriate threshold of the QR location information. Then, we can get four accurate vertexes coordinates of QR code image and accomplish the accurate correction of QR code image based on projection transformation. The algorithm is able to complete effective correction for these captured images under different environments and solve the key technical bottlenecks of QR code recognition.
Vision systems have been used in many applications that intends to reduce the need for human operators. This is especially true for tasks that are simple but repetitive in nature, which is largely applicable to most m...
Vision systems have been used in many applications that intends to reduce the need for human operators. This is especially true for tasks that are simple but repetitive in nature, which is largely applicable to most manufacturing and agriculture's post-harvest processes. Many such processes utilize conveyor-based systems where the objects being processed are placed on a conveyor belt that runs through multiple processing stations. Implementing a vision system to capture images of an object that is moving usually requires setting up an imaging device to a working conveyor system. Getting a working conveyor system to be ready can take some time and consequently delay development work on the vision system itself, especially those involving imageprocessingalgorithms. This paper proposes a software solution that can be used to expedite initial work on such systems. The solution is written in C and is therefore easily ported to any development machine. A basic imageprocessing library has also been developed so that it does not depend on any development library or suite, which is usually huge in size. Thus, the solution can easily be compiled and run on embedded development boards like Raspberry Pi - for a more portable solution.
A biometric system acquires biometric features from an individual and compare these features with other features stored in the database. Iris recognition system is a reliable and an accurate biometric system. Localiza...
详细信息
A biometric system acquires biometric features from an individual and compare these features with other features stored in the database. Iris recognition system is a reliable and an accurate biometric system. Localization of the iris boundaries in an eye image is considered to be the most vital step in the iris recognition process. There exist many algorithms to segment the iris. One of the segmentation methods, that is used in many commercial iris biometric systems is an algorithm known as a Daugman's algorithm. Especially it focuses on image segmentation and feature extraction for iris recognition process but Daugman uses more processing time. This paper implements the proposed algorithm to achieve best performance in terms of accuracy and time. The implemented algorithm was tested on UBIRIS V.1 database which includes 15 individuals from both Right and Left eyes resulting in 45 classes in total. The proposed algorithm attains an overall accuracy of 95% with robust performance.
Parallelization on a GPU (graphics processing unit) cluster is an effective approach to reducing the huge computation time of backprojection, which is the most accurate SAR (synthetic aperture radar) imaging algorithm...
详细信息
ISBN:
(纸本)9781538623268
Parallelization on a GPU (graphics processing unit) cluster is an effective approach to reducing the huge computation time of backprojection, which is the most accurate SAR (synthetic aperture radar) imaging algorithm for reconstructing images with no errors caused by the platform motion. To obtain accurate imagery in real-time, we developed a distributed parallel backprojection algorithm for stripmap SAR on GPU clusters, which reconstruct the image while receiving signals from the remote platform. In the case of receiving the 1.9 GiB signals from the remote storage through 1GbE, we found that 16 GPUs on the 4 nodes are 11.5 times faster than 1 GPU, and they finished the imaging 1.0s after receiving all signals.
Digital images and digital imageprocessing facilitated significant progress in numerous areas where medicine is an important one of them. Computer-aided detection and diagnostics systems are used to assist specialist...
详细信息
ISBN:
(纸本)9788086943503
Digital images and digital imageprocessing facilitated significant progress in numerous areas where medicine is an important one of them. Computer-aided detection and diagnostics systems are used to assist specialists in interpretation of medical digital images. One of the important research issues is detection and classification of the chronic obstructive pulmonary disease in lung CT images In this paper we proposed a method for emphysema classification based on texture and intensity features. Only six different characteristics of the uniform local binary pattern and intensity histogram were used as input vector for support vector machine that was used as classifier. Feature vector was significantly reduced compared to the other state-of-the-art methods while the classification accuracy was increased. On images from standard dataset global accuracy of our proposed algorithm was 98.18% compared to 95.24% and 93.9% of two other compared algorithms.
暂无评论