The article presents a method of recognizing alphanumeric characters located in the image, based on a previously created database of patterns using neural networks. For this purpose the convolutional networks were used, which independently search for features that allow to distinguish characters in the image. A larger number of convolution layers allows us to recognize a greater number of features and thus to increase the probability of correctly recognized characters. The main purpose of the paper is to present software that recognizes the alphanumeric characters in images and to investigate the impact of the size of this database on the program's speed and character recognition efficiency. This software can also be used in more complex structures, such as automatic translators or as a computer reader. The calculation of the first program that recognizes single character and the second program that reads all the text from the image have been made in the MATLAB environment. The paper describes the components of this software, such as the learning subsystem and the character recognition subsystem. The results of the program were presented in the form of screenshots showing the results of the learning process and character recognition process. The speed of the software and the effectiveness of recognizing alphanumeric characters using the artificial neural networks and maximally stable extremal regions (MSER) algorithm are presented in the table and figures. Attention was also paid to the impact of the size of the database used to learn the network on the speed of calculations and recognition efficiency.
The article presents a way of using machine learning algorithms to recognize objects in images. To implement this task, an artificial neural network was used, which has a high adaptability and allows work with a very large set of input data. The neural network was described using a program written in the MATLAB simulation environment. The basic problem faced by the designer of objects recognition is to collect a sufficient training set of images to achieve the high probability of correct recognition. The set of learning patterns in the artificial neural networks may contain from several dozen thousands to one million training samples. In this article at the beginning the neural network was pre-trained trained based on the images included in the publicly available CIFAR 100 database, which are characterized by a small size of 32x32 pixels. It contains 70 000 images assigned to 10 basic categories. Then the author's database, consisting from 1000 pedestrians, cars and road signs was used. The article contains a description of applied algorithm, method of supervised learning and correction of weight coefficients, selection of activation function and operation on max pooling filter. The results of proposed solution are presented in the form of screenshots from calculations and in figures depicting results of recognized objects. Attention was also paid to the impact of used database for learning the network on the speed of calculations and recognition efficiency. The proper selection of number and types of layers, number of neurons, activation function and the value of the learning factor is very important in designing the neural network in application to objects recognition contained in the images. The problems occurring in the process of learning the neural networks and suggestions for their further improvement are also presented.
KEYWORDS: Signal processing, Neural networks, Signal generators, Digital signal processing, Electromagnetism, Detection and tracking algorithms, Radar, Databases, Signal to noise ratio, Electronics
The article presents a method of recognizing sources of electromagnetic signal emission on the basis of signals generated by using deep neural networks. These signals are measured in electronic recognition receivers, processed into a digital signal and then undergo recognition. The main purpose of the article is to present software which, based on the detected signal, is to recognize it. The software can also be used as a subsystem in Electronic Intelligence (ELINT) devices, including detection of radiolocation systems, jammers, recognition of aircrafts, ships, vehicles based on the signal shape of radar cross section (RCS) and subsequently comparison them to the emitter database (EDB). The implementation of this system is presented in a simulation environment and with the help of a signal generator that has the ability to make changes in signal signatures earlier recognized, calculated and written in database. The proposed software allows to examine a significant number of different signals. The article contains a description of components of software, such as signal base, learning subsystem and signal generator. The results of the system operation are presented in the form of screenshots from individual software components. The speed of software operation, the effectiveness of recognition systems using artificial neural networks is presented by means of tables and appropriate illustrations. Also presented is the problem of learning the neural networks at the GPUs (graphics processing units) and the way of choice the learning coefficients.
The article presents a short description of the methods of shielding terrestrial objects using land jammers. The necessary mathematical formulae needed to calculate the effectiveness of radar jamming to shield a single ground object for two variants are given: when the jammer is near to the shielded object and when it is away from it at a fixed distance. For the energy criterion, the method for calculating the sector of the ground object shield was presented, according to the parameters of the jammer, jammed airborne radar and aircraft flight altitude. The necessary mathematical relations have been given to determine the sector of the object's shield. The angular relations and distances between the object to be covered, the jammers and the attacking aircraft change during the raid. The results of numerical calculations illustrating the presented method for the exemplary parameters of the jammer station, the radar and the variant of the air raid are presented in the respective tables and diagram. The implementation of an effective object shield of the facility using land-based jammers depends mainly on the exact knowledge of the technical parameters of the jammed radars (the emitter database), the jammers, the distance between them and the attacking aircraft, the degradation coefficient used for calculations, the electromagnetic wave propagation conditions and the used jamming method. In order to achieve high effectiveness of on-board radars jamming, the software supporting the decision making process depends on the dynamically changing battle scenarios.
KEYWORDS: Radar, Electronics, Warfare, Signal detection, Electromagnetism, Receivers, Signal processing, Antennas, Electronic components, Signal attenuation
The electronic countermeasures (ECM) include methods to completely prevent or restrict the effective use of the electromagnetic spectrum by the opponent. The most widespread means of disorganizing the operation of electronic devices is to create active and passive radio-electronic jamming. The paper presents the way of jamming efficiency calculations for protecting ground objects against the radars mounted on the airborne platforms. The basic mathematical formulas for calculating the efficiency of active radar jamming are presented. The numerical calculations for ground object protection are made for two different electronic warfare scenarios: the jammer is placed very closely and in a determined distance from the protecting object. The results of these calculations are presented in the appropriate figures showing the minimal distance of effective jamming. The realization of effective radar jamming in electronic warfare systems depends mainly on the precise knowledge of radar and the jammer’s technical parameters, the distance between them, the assumed value of the degradation coefficient, the conditions of electromagnetic energy propagation and the applied jamming method. The conclusions from these calculations facilitate making a decision regarding how jamming should be conducted to achieve high efficiency during the electronic warfare training.
The paper describes some of the problems associated with emitter location calculations. This aspect is the most important part of the series of tasks in the electronic recognition systems. The basic tasks include: detection of emission of electromagnetic signals, tracking (determining the direction of emitter sources), signal analysis in order to classify different emitter types and the identification of the sources of emission of the same type. The paper presents a brief description of the main methods of emitter localization and the basic mathematical formulae for calculating their location. The errors’ estimation has been made to determine the emitter location for three different methods and different scenarios of emitters and direction finding (DF) sensors deployment in the electromagnetic environment. The emitter has been established using a special computer program. On the basis of extensive numerical calculations, the evaluation of precise emitter location in the recognition systems for different configuration alignment of bearing devices and emitter was conducted. The calculations which have been made based on the simulated data for different methods of location are presented in the figures and respective tables. The obtained results demonstrate that calculation of the precise emitter location depends on: the number of DF sensors, the distances between emitter and DF sensors, their mutual location in the reconnaissance area and bearing errors. The precise emitter location varies depending on the number of obtained bearings. The higher the number of bearings, the better the accuracy of calculated emitter location in spite of relatively high bearing errors for each DF sensor.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.