The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The aim of this paper is to present a comparative study of a set of neural network based controlling methods for the pitch angle of an aircraft. The investigated methods can be classified in two categories: the first one is based on direct and inverse offline neural modelling of the process, while the second one is based on the online training of a neural controller and on the online self-tuning of...
Text segmentation is an important problem in document analysis related applications. We address the problem of classifying connected components of a document image as text or non-text. Inspired from previous works in the literature, besides common size and shape related features extracted from the components, we also consider component images, without and with context information, as inputs of the...
Computer graphics techniques for image generation are living an era where, day after day, the quality of produced content is impressing even the more skeptical viewer. Although it is a great advance for industries like games and movies, it can become a real problem when the application of such techniques is applied for the production of fake images. In this paper we propose a new approach for computer...
Urban traffic flow is dynamic and uncertain. In this paper, we combine the deep learning and the reinforcement learning, and design an intersection signal controller based on Q-learning and convolutional neural network. We redefine the state space and the reward function. The training and simulation of the controller are carried out in traffic micro-simulator SUMO. Compared with timing control, the...
Image dehazing can be described as the problem of mapping from a hazy image to a haze-free image. Most approaches to this problem use physical models based on simplifications and priors. In this work we demonstrate that a convolutional neural network with a deep architecture and a large image database is able to learn the entire process of dehazing, without the need to adjust parameters, resulting...
Localization is one of the main tasks of autonomous mobile robots. There are many approaches on how to determine the robot's position with reasonable precision, as with the use of sensor fusion (IMU, GPS, Image, LiDAR). Even though it is possible to achieve high precision with these sensors combined, a solution that requires less resources in terms of processing, energy consumption and yet provide...
Deep learning is nowadays one of the most popular research topics in computer science. In recent years, the extensive application of convolutional neural network has made it become a new direction for the computer architecture research that is developing rapidly. Currently, there is a growing demand on off-line deploying deep learning network on top of embedded mobile systems. However, how to balance...
New Renewable Energy sources are changing the way the Electric Grid is conceived: new challenges are given to Distribution Network Operators (DNOs) and to Transmission Systems Operators (TSOs). These challenges regards business models adopted and reliability, efficiency and controllability of the system. In this new framework, it is important to give to Decision Makers tools able to help them in facing...
Accurate prediction of the future locations of the host vehicle as well as that of the surrounding objects is one of the key challenges in improving road traffic safety. The traditional approach for this task has been using physics-based motion models such as kinematic and dynamic models, the result of which is not reliable for long-term prediction. In this paper, we present simulation results demonstrating...
Deep Denoising Autoencoder (DDAE) is an effective method for noise reduction and speech enhancement. However, a single DDAE with a fixed number of frames for neural network input cannot extract contextual information sufficiently. It has also less generalization in unknown SNRs (signal-to-noise-ratio) and the enhanced output has some residual noise. In this paper, we use a modular model in which three...
Software fault prediction is one of the significant stages in the software testing process. At this stage, the probability of fault occurrence is predicted based on the documented information of the software systems that are already tested. Using this prior knowledge, developers and testing teams can better manage the testing process. There are many efforts in the field of machine learning to solve...
Most present methods of saliency detection emphasize too much on the local contrast while ignore the global feature of image. The detailed characteristics of the image can be reflected based on the local comparison of image. However, the overall saliency of the image cannot be reflected. In this paper, a saliency detection model combined local and global features was proposed. Firstly, a local feature...
We propose associative domain adaptation, a novel technique for end-to-end domain adaptation with neural networks, the task of inferring class labels for an unlabeled target domain based on the statistical properties of a labeled source domain. Our training scheme follows the paradigm that in order to effectively derive class labels for the target domain, a network should produce statistically domain...
Input binarization has shown to be an effective way for network acceleration. However, previous binarization scheme could be regarded as simple pixel-wise thresholding operations (i.e., order-one approximation) and suffers a big accuracy loss. In this paper, we propose a highorder binarization scheme, which achieves more accurate approximation while still possesses the advantage of binary operation...
Riding on the waves of deep neural networks, deep metric learning has achieved promising results in various tasks by using triplet network or Siamese network. Though the basic goal of making images from the same category closer than the ones from different categories is intuitive, it is hard to optimize the objective directly due to the quadratic or cubic sample size. Hard example mining is widely...
Understanding the visual relationship between two objects involves identifying the subject, the object, and a predicate relating them. We leverage the strong correlations between the predicate and the hsubj; obji pair (both semantically and spatially) to predict predicates conditioned on the subjects and the objects. Modeling the three entities jointly more accurately reflects their relationships...
Binaural features of interaural level difference and interaural phase difference have proved to be very effective in training deep neural networks (DNNs), to generate time-frequency masks for target speech extraction in speech-speech mixtures. However, effectiveness of binaural features is reduced in more common speech-noise scenarios, since the noise may over-shadow the speech in adverse conditions...
It is a simple task for humans to visually identify objects. However, computer-based image recognition remains challenging. In this paper we describe an approach for image recognition with specific focus on automated recognition of plants and flowers. The approach taken utilizes deep learning capabilities and unlike other approaches that focus on static images for feature classification, we utilize...
This work presents improvements to a neuroevolution algorithm called Evolutionary eXploration of Augmenting Convolutional Topologies (EXACT), which is capable of evolving the structure of convolutional neural networks (CNNs). While EXACT has multithreaded and parallel implementations, it has also been implemented as part of a volunteer computing project at the Citizen Science Grid to provide truly...
We propose a gray coding method for deep neural network (DNN) based decoder. With multiple resources considered together, DNN can be used to decode corrupted signals. In deep learning training, stochastic gradient descent (SGD) algorithm is used, which means that the cost function must be differentiable. Then, allocating the discrete bits for each symbol is difficult. To solve this problem, the basic...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.