• 2015-2024close
• Open Access close
• Closed Access close
• Restricted close
• Publications close
• Research data close
• Other research products close
• CZ close
• English close
• Digitální knihovna VUT close
• North American Studies close

In this paper, we present two pipelines in order to reduce the feature space for anomalydetection using the One Class SVM. As a first stage of both pipelines, we compare the performanceof three convolutional autoencoders. We use the PCA method together with t-SNE as the first pipelineand the reconstruction errors based method as the second. Both methods have potential for theanomaly detection, but the reconstruction error metrics prove to be more robust for this task. Weshow that the convolutional autoencoder architecture doesn’t have a significant effect for this task andwe prove the potential of our approach on the real world dataset.

We propose the model combining convolutional neural network with multiple instancelearning in order to localize the premature atrial contraction and premature ventricular contraction.The model is based on ResNet architecture modified for 1D signal processing. Model was trainedon China Physiological Signal Challenge 2018 database extended by manually labeled ground truthpositions of premature complexes. The presented method did not reach satisfying results in PAClocalization (with dice = 0.127 for avg-pooling implementation). On the other hand, results of localizationof PVCs were comparable with other published studies (with dice = 0.952 for avg-poolingimplementation).

SILON s.r.o is manufacturer of polyester fibres which get used in wide range of applications, many of them requiring highest quality material. Due to manufacturing processes, some fibres are not drawn properly and stay in the fiber as bundles, or brittle, thick threads. Proposed lab station should automate process of quality check of each batch. It consists of linescan camera scanner and computer with software for detection and analysis of defects.

This work deals with the use of a convolutional neural network in the area of segmentation of images acquired with the use of a transmission electron microscope. Paper describes programming tool for image data augmentation, used neural network topology, and it also provides information about model training. This neural network topology delivered excellent results on provided data from the Thermo Fisher Scientific company, which will serve as a starting point for internal company research in image segmentation area.

To protect the vineyards from starlings is very costly and ineffective with the available resources. In my work, I have designed a detection module which consist of four microphones, each one is directed to the cardinal point and based on the intensity of sound, the module will provide the information about the direction where the flock is located, in the vineyard that passes to the control module and scares the flocks. The detection module processes the signal in a Raspberry Pi 4 single board computer using an artificial neural network algorithm.

This paper deals with proposed 2nd order universal frequency filter. In design of this frequency filter, multiple-output current follower (MO-CF) and digitally adjustable current amplifier (DACA) have been used. The frequency filter is proposed with ability to control the pole frequency without disturbing the selected quality factor by parameter of DACA active elements in form of integrated circuit. Design is supported by simulation and also measurement results.

In 1939, due to WWII and the Nuremberg Laws, the revolutionary Czech structural engineer Jaroslav J. Polívka arrived in the United States. After his arrival, he started a research job at UC Berkeley, renewed his engineering practice, and offered his services to the US military as many businesspersons did during this era. Polívka worked for Henry Kaiser who turned Richmond, CA, into a vibrant, fast developing workers city. New residential districts, hospitals, hangars, docks, and warehouses were built there. Henry Kaiser approached the structural development of the city in the same way he revolutionized the construction of battleships: from prefabricated, standardized parts. He supported research and development of new technologies. Mobile, round-shaped hospitals from prefabricated aluminum frames were one of the results of that research. In 1946, Jaroslav J. Polívka introduced himself to the “starchitect” Frank Lloyd Wright. A productive mutual co-operation that lasted 13 years and resulted in eight spectacular projects had started and Polívka, who had been working on extensive research both at UC Berkeley and Stanford University, came up with many technological, structural, and material innovations over the period. In 1957, Henry Kaiser funded the construction of one of the two geodesic domes designed by Richard Buckminster Fuller in Hawaii and in the process he invited Frank Lloyd Wright to consult the project. Jaroslav J. Polívka was probably in direct contact with Fuller, since he wanted to include him in his unfinished project of an encyclopedia of the world-famous structural engineers. On this particular story and a social matrix evolving around Henry Kaiser and Frank Lloyd Wright, the lecture seeks to rethink architectural global modernism as a cooperative project rather than a series of individual innovations manifested by isolated genius figures.

A new fully-differential (F-D) filtering structure of the 2nd-order current-mode universal filter is described in this paper. Presented filter is also proposed in a single-ended (S-E) form in order to compare behavior of single-ended and fully-differential structures. Signal-flow graphs method is used to design both presented filters. Multi-output transconductance amplifiers (MOTAs), digitally adjustable current amplifier (DACA), multi-output current follower (MO-CF) and fully differential current follower (FD-CF) were used in proposal. Filters possess of ability to adjust their pole frequency and quality factor independently of each other. The actual function of proposed filters has been verified by PSpice simulations and experimental measurements.

The thesis deals with the image processing and detection of specific objects in the image. The main objective of this work is to implement a algoritm for the control of pipetting processes based on images from the camera. In this thesis is used for image processing an open source computer vision library OpenCV and for pipette detection is used convolutional neural network. The proposed solution is still in development process. The following article describes the issue and the results of the thesis solution.

Image similarity measuring has many various applications. Pedestrian recognition is one of them and for the security purposes it is basically required to run in real-time. This paper proposes a deep Siamese neural network architecture for pedestrian recognition that achieves 70.28% accuracy on the test set containing 20 persons. Prediction of the model is fast enough for real-time processing.