Search

Search Results

• THE METRICS FOR TRACKING ALGORITHMS EVALUATION

  А. Е. Shchelkunov, V.V. Kovalev, K.I. Morev, I.V. Sidko

  2020-07-10

  Abstract ▼

  The work is devoted to a review of existing metrics for assessing the quality of the task of tracking objects on video with various algorithms. When evaluating tracking algorithms for their subsequent com-parison, it is not enough to use one metric, and algorithms should be evaluated using a set of different independent estimates. To this end, a study was conducted of existing metrics for evaluating algorithms, the results of which are given in the article. The review involves many different approaches to evaluating algorithms. For example, approaches based on the assessment of the definition of the center of the track-ing object, which are one of the first and still popular metrics for evaluating tracking algorithms. The main disadvantages of such approaches include the difficulty of determining the true center of the object, as well as the interpretation of estimates for various sizes of the object. To eliminate these shortcomings, anew metric is introduced in the article: an unbiased (window) error in determining the center of an object, which takes into account the constant component of the error in determining the center. Other approach-es include metrics based on the analysis of the intersection over union. Also, the article considers ap-proaches based on the analysis of tracking failures, which take into account the tracking length and fail-ure rate. A new method is proposed for evaluating algorithms in case of loss of visual contact with an tracking object, taking into account the number of frames in which visual contact with the object was lost. During the study, approaches to evaluating algorithms for simultaneous tracking of several objects were considered. Integral metrics were proposed whose task is to obtain a comprehensive assessment of the tracking algorithm. For the formation of a comprehensive assessment, it is desirable to use various un-correlated metrics. Complex estimates provide the ability to compare algorithms with each other. As a comprehensive assessment, the article proposes the use of a metric combining the accuracy and robust-ness of the algorithm. As a rule, the intersection over union is used as the accuracy metric, however, for problems where the accuracy of tracking the center of the object is fundamental, the authors propose using an unbiased error in determining the center as the accuracy metric.

• MULTI-AGENT ALGORITHM FOR AUTOMATIC DETECTION AND TRACKING OF NON-DETERMINISTIC OBJECTS

  V.A. Tupikov, V.A. Pavlova, V.A. Bondarenko, A.I. Lizin, D.K. Eltsova, M.V. Sozinova

  2020-07-10

  Abstract ▼

  In order to develop a robust algorithm for the automatic detection and tracking of non-deterministic objects for embedded computing systems, in this work, a study and analysis in the field of state-of-the-art general-purpose automatic tracking algorithms is performed. The most successful of those algorithms suitable for long-term stable automatic tracking of objects (without a priori knowledge of the type of object being tracked) have already gone beyond solving exclu-sively tracking problems, and include a synergistic combination of several heterogeneous tracking algorithms, as well as at least one automatic detection and / or classification algorithm. Thus, the authors of the article conclude that the most stable modern automatic tracking algorithms are a multi-agent system that makes a decision about the current position, size and other parameters of the tracked object image based on intelligent voting of system’s submodules that independently monitor the object and form its model. Individual models of each of the submodules are updated based on the results of a collective decision. The authors of the study identified the most effective of the applied basic algorithms suitable for use in embedded computing systems of robotic systems, and developed a new multi-agent algorithm for the automatic detection and tracking of non-deterministic objects. The presented multi-agent algorithm includes a submodule for extracting and matching key points in images, a clustering and filtering submodule for key points using the DBSCAN algorithm, a tracking submodule based on the optical flow calculation algorithm, and a key point classification submodule. A semi-natural testing of the developed algorithm was carried out and its effectiveness in solving tasks not only of automatic tracking of objects, but also in tasks of automatic objects detection using several reference images were evaluated. In conclusion, the authors present steps for further improving the accuracy and performance of the developed algo-rithm for its forthcoming implementation for on-board computing systems of aerial vehicles.

• OBJECT DETECTION ALGORITHM FOR OPTOELECTRONIC SYSTEMS WITH ONLINE LEARNING

  V.A. Tupikov, V.A. Pavlova, V.A. Bondarenko, M.V. Sozinova, P.A. Gessen

  2021-04-04

  Abstract ▼

  In order to create a new algorithm for automatic detection of objects with real-time training, a
  study of the world scientific groundwork in the field of general-purpose automatic tracking with the
  ability to recognize a tracked object with the potential for application in embedded computing systems
  of optoelectronic systems of promising robotic complexes was carried out. Based on the conducted
  research, methods and approaches were selected and tested that allow, with the greatest accuracy,
  while maintaining high computational efficiency, to provide training of classifiers on the fly
  (online learning) without a priori knowledge of the type of tracking object and to ensure the subsequent
  detection of the original object in the event of its short-term loss. Such methods include a histogram
  of oriented gradients – a descriptor of key features based on the analysis of the distribution of
  the brightness gradients of the object image. Its use allows you to reduce the amount of information
  used without losing key data about the object and to increase the speed of image processing. The
  article substantiates the choice of one of the real-time classification algorithms that allows solving
  the problem of binary classification – the support vector machine. Due to the high speed of data processing
  and the need for a small amount of initial training data to construct a separating hyperplane,
  on the basis of which the classification of objects is done, this method is chosen as the most suitable
  for solving the problem. For online training, a modification of the support vector machine method
  was chosen, which implements stochastic gradient descent at each step of the algorithm – Pegasos.
  The authors of the study carried out the development and semi-natural modeling of the selected algorithm,
  evaluated the effectiveness of its work in the tasks of detecting an object of interest in real time
  with preliminary online training in the process of tracking the object. The developed algorithm has
  shown high efficiency in solving the problem and is planned to be implemented as part of a special
  software for optoelectronic systems of advanced robotic systems. In the conclusion, proposals are
  presented to further improve the accuracy and probability of the object detection by the developed
  algorithm, as well as for improving its performance by optimizing calculations.

• MODULE FOR ADJUSTING PARAMETERS OF ALGORITHMS FOR AUTOMATIC DETECTION AND TRACKING OF OBJECTS FOR OPTOELECTRONIC SYSTEMS

  V. А. Tupikov, V. А. Pavlova, А.I. Lizin, P.А. Gessen

  71-81

  2022-04-20

  Abstract ▼

  In order to create an innovative module for automatic correction of algorithms for automatic
  detection and tracking of objects with real-time training, a study of world experience in the field
  of general-purpose automatic tracking with the ability to recognize the tracking object for use in
  embedded computing devices of optoelectronic systems of promising robotic complexes was carried
  out. Based on the conducted research, methods and approaches have been selected and tested
  that allow with the greatest accuracy, while maintaining high computational efficiency, to provide
  on-the-fly training of classifiers (online learning) without a priori knowledge of the type of tracking object and to ensure subsequent correction during tracking and detection of the original object
  in case of its short-term loss. Such methods include a histogram of directional gradients – a descriptor
  of key features based on the analysis of the distribution of brightness gradients of an object
  image. Its use allows you to reduce the amount of information used without losing key data
  about the object and increase the speed of image processing. The article substantiates the choice
  of one of the classification algorithms in real time, which allows solving the problem of binary
  classification - the method of support vectors. Due to the high speed of data processing and the
  need for a small amount of initial training data to build a separating hyperplane, on the basis of
  which the classification of objects takes place, this method is chosen as the most suitable for solving
  the task. To implement online training, a modification of the support vector machine was chosen,
  implementing stochastic gradient descent at each step of the algorithm – Pegasos. Another
  auxiliary method is the clustering method of key points – this ensures an accelerated selection of
  objects for classification and training. The authors of the study carried out the development and
  semi-natural modeling of the proposed module, evaluated the effectiveness of its work in the tasks
  of correcting and detecting the object of interest in real time with preliminary online training in
  the process of tracking the object. The developed algorithm has shown high efficiency in solving
  the problem. In conclusion, proposals are presented to further improve the accuracy and probability
  of detecting an object of interest by the developed algorithm, as well as to improve its performance
  by optimizing calculations.

• PHASE TRACKING LOOPS SUPPORTING IN THE SATELLITE NAVIGATION RECEIVER USING INERTIAL NAVIGATION SYSTEM MEASUREMENTS

  А.А. Cherkasova, А. Y. Shatilov, Т.А. Mukhamedzyanov

  2022-04-21

  Abstract ▼

  Satellite radio navigation systems make it possible to evaluate the user's state vector, use
  coordinates, user speed and time relative to the system scale. The requirements for the characteristics
  of these systems constantly depend on the fact that they have application features in their
  algorithms for processing radio navigation signals. One of the main characteristics of satellite
  radio navigation systems is the accuracy of estimating the user's state vector. This characteristic
  can be improved by the presence of estimates of the phase of the received radio navigation signals.
  In a satellite radio navigation system, phase estimation errors in the tracking loop have two components:
  dynamic and noise. To compensate for the noise error, it is necessary to reduce the
  equivalent noise band of the anti-aliasing filter of the phase tracking loop. However, the minimumpossible bandwidth of the smoothing filter is limited by the presence of consumer dynamics and the
  quality of the reference oscillator. As a result, in the presence of consumer dynamics, the sensitivity
  and reliability of phase tracking deteriorates. To compensate for the dynamic error in the phase
  tracking loop, information from an inertial navigation system can be used. The satellite radio navigation
  system and the inertial navigation system have complementary characteristics. The use of
  support for phase tracking loops from an inertial navigation system makes it possible to increase
  the sensitivity and reliability of its operation in the presence of consumer dynamics. It is assumed
  that with such an implementation, the sensitivity of the phase tracking loops will be limited only by
  the instability of the reference oscillator and the error of inertial measurements. To improve the
  characteristics of accuracy, sensitivity and reliability of the coherent mode of operation of the end
  device, an algorithm was developed to support phase tracking loops with measurements from an
  inertial navigation system. A study of the developed algorithm was carried out on a model that
  uses real measurements of satellite and inertial navigation systems as input data. The developed
  algorithm is implemented in the software of the NV216C-IMU inertial satellite navigation system
  prototype. Experimental studies were carried out in the conditions of automobile dynamics in open
  areas. The research results are presented in the work.

• CORRELATIONAL SUPPORT ALGORITHM WITH REAL-TIME LEARNING

  V. А. Tupikov, V. А. Pavlova, А.Y. Gagarina, P. А. Gessen, А.I. Lizin, М. V. Sozinova

  2022-04-21

  Abstract ▼

  In order to develop a stable algorithm for automatic detection and tracking of nondeterministic
  objects with real-time learning for embedded computing systems with optoelectronic
  devices, within the framework of this work, a study and analysis of the existing world scientific and
  technical experience in the field of automatic tracking algorithms for general purposes was carried
  out. The article shows that the most stable modern automatic tracking algorithms are a system
  that makes a decision about the current position, size and other parameters of the tracked
  image based on the model being trained. The authors of the study identified the most effective of
  the applied basic algorithms suitable for use in embedded computing systems of robotic complexes,
  and developed a new algorithm for automatic detection and maintenance of non-deterministic
  objects. A semi-natural testing of the developed algorithm was carried out and its effectiveness
  was evaluated in solving problems not only of automatic tracking of objects, but also problems of
  automatic detection of objects using several reference images. In conclusion, proposals are presented
  for further improving the accuracy of the developed algorithm and for its optimization and
  implementation in the special software of on-board computer systems of aircraft.

• ACCELERATION OF THE DIRECT PASSAGE IN THE IMPLEMENTATION OF CNN ON A LIMITED COMPUTING RESOURCE

  А.Е. Shchelkunov, V.V. Kovalev, I. V. Sidko, N. Е. Sergeev

  2022-04-21

  Abstract ▼

  The work is devoted to the optimization of the neural network architecture for its launch on
  a limited computing resource. Several optimization approaches are considered, estimates of the
  complexity and execution time of the forward pass of the neural network are given. Comparative
  estimates of the complexity of the network using different optimization approaches are given.
  The paper presents an analysis of the selected network architecture, and estimates of the computational
  complexity of individual components (modules) of the architecture are obtained. An analysis
  of possible optimization methods for each module was made. The parameters of the considered
  modules, the sizes of the input and output tensors are described. Several architectures were tested
  to optimize the feature extraction module, ResNet 50, ResNet 18, MobileNet v3 small, MobileNet
  v3 large. A comparative analysis of the computational complexity and execution time of the forward
  pass for each architecture is presented. Forward pass times were measured on Nvidia's
  Jetson AGX Xaver embedded computing device. Estimates of the execution time of the direct pass
  for each module of the considered neural networks are presented. The paper presents the results of
  comparing neural network accuracy estimates before and after architecture optimization. The test
  data set consists of 100 video recordings. 5 different typical objects are involved in test videos,
  10 different scenarios are recorded for each object class. For each of the developed architectures,
  accuracy estimates were obtained, and a comparative analysis was made.

• MULTYCHANEL ADAPTIVE PHASE LOCK LOOP SYSTEM FOR GNSS RECEIVER

  А. А. Cherkasova, А. Y. Shatilov

  2025-04-27

  Abstract ▼

  Satellite navigation equipment often operates under conditions of a priori uncertainty of the parameters
  of the mutual dynamics between the transmitter and the consumer and the signal-to-noise ratio of the
  received signals of satellite radio navigation systems. Classical Bayesian algorithms for Phase lock loop
  system require a priori knowledge about the parameters of the phase process dynamics and the signal-tonoise
  ratio (SNR) of the received signals. As a result, the operation of such algorithms under conditions
  other than that a priori specified is not optimal to the criterion of minimum error variance. Moreover, a
  sudden change in the signal-to-noise ratio or dynamics can lead to a tracking failure in such a system.
  The purpose of this work is to develop an optimal phase tracking system that is adaptive to the dynamics
  of the phase process and the signal-to-noise ratio in order to maintain phase tracking in the widest possible
  range of operating conditions while tracking global navigation satellite system signal. An adaptive
  multichannel phase lock loop system has been synthesized as a result of formulation and solution of signal
  processing problem in terms of the statistical synthesis theory. Adaptivity to the changing power of the
  received signal is achieved by including the signal-to-noise ratio [dBHz] in the vector of estimated filter
  parameters. Adaptability to the intensity of the phase change dynamics is achieved through the use of a
  multi-channel filtration system. Statistical modeling of an adaptive multichannel phase lock loop tracking
  system with a complex algorithm for tracking the code delay of the signal of satellite radio navigation
  systems has been carried out. The values of the sensitivity of phase tracking under various dynamic conditions
  are determined. The adaptive multichannel phase lock loop system is able to withstand an signal-tonoise
  ratio jump from 50 to 10 dBHz and back without loss of phase tracking in low dynamics conditions
  (only the drift of the reference quartz oscillator). The AMPLL system is able to withstand abrupt transitions
  of dynamics between low and high (the sinusoidal acceleration 10g and sinusoidal jerk 10 g/s) without
  loss of phase tracking under the 24 dBHz signal-to-noise ratio. Thus, in real conditions, when the dynamics
  of the GNSS receiver and the SNRs of the received signals change in an unpredictable way, the
  AMPLL system keep tracking in a much wider range of conditions than the non-adaptive PLL