Image Colorization

The article discusses one of the latest ways to colorize a black and white image using deep learning methods. For colorization, a convolutional neural network with a large number of layers (Deep convolutional) is used, the architecture of which includes a ResNet model. This model was pre-trained on images of the ImageNet dataset. A neural network receives a black and white image and returns a colorized color. Since, due to the characteristics of ResNet, an input multiple of 255 is received, a program was written that, using frames, enlarges the image for the required size. During the operation of the neural network, the CIE Lab color model is used, which allows to separate the black and white component of the image from the color. For training the neural network, the Place 365 dataset was used, containing 365 different classes, such as animals, landscape elements, people, and so on. The training was carried out on the Nvidia GTX 1080 video card. The result was a trained neural network capable of colorizing images of any size and format. As example we had a speed of 0.08 seconds and an image of 256 by 256 pixels in size. In connection with the concept of the dataset used for training, the resulting model is focused on the recognition of natural landscapes and urban areas.

3D-Cell-Annotator: an open-source active surface tool for single cell segmentation in 3D microscopy images

SummarySegmentation of single cells in microscopy images is one of the major challenges in computational biology. It is the first step of most bioimage analysis tasks, and essential to create training sets for more advanced deep learning approaches. Here, we propose 3D-Cell-Annotator to solve this task using 3D active surfaces together with shape descriptors as prior information in a fully- and semi-automated fashion. The software uses the convenient 3D interface of the widely used Medical Imaging Interaction Toolkit (MITK). Results on 3D biological structures (e.g. spheroids, organoids, embryos) show that the precision of the segmentation reaches the level of a human expert.Availability and implementation3D-Cell-Annotator is implemented in CUDA/C++ as a patch for the segmentation module of MITK. The 3D-Cell-Annotator enabled MITK distribution can be downloaded at: www.3D-cell-annotator.org. It works under Windows 64-bit systems and recent Linux distributions even on a consumer level laptop with a CUDA-enabled video card using recent NVIDIA [email protected] and [email protected]

Mathematical simulation of aircraft engine jet exhausts radiation

The paper deals with the problems of mathematical simulation of aircraft engine jet exhausts radiation, the simulation being carried out by means of shader subroutines for the concurrent computation of the radiative transfer equation on the video card resources. The combination of an analytical model of an isobaric jet and ray tracing of computation of the radiative transfer equation allows us to develop a flexible model of aircraft jet radiation, the model taking into account the main parameters of streams in the jet and in the co-current flow, the spectral lines of the radiating components, and provides real-time computation. For the graphic implementation of the model, the OpenGL standard is used

PROSPECTS OF THE MULTI-PURPOSE COMPUTER PLATFORM OPPORTUNITIES DEVELOPMENT

In the article possibilities of increase of speed for solving a certain class of problems with the help of a profit center created in NIIVK named after М. А. Kartsev. As such capabilities are considered video cards and an optical module InLight256, the dimensions of which allow it to be included in the structure of the profit center. As an example of the use of video cards to improve the speed of implementation of algorithms, an algorithm is considered for finding the shortest coverings of Boolean matrices and further use of the results of this algorithm to find the minimal forms of weakly defined Boolean functions. The article shows that the use of a video card for the class of problems under consideration makes it possible to obtain a gain in time several times. The optical module InLight256, developed by the Israeli company Lenslet, is currently the only optical processor that is commercially available, which can be purchased. The module is equipped with software and is compatible with modern electronic modules. The article lists the advantages of optical modules in comparison with electronic modules and the main characteristics of the above module. In addition, the possible areas of application of the module for both peaceful purposes and military are indicated.

CUDAICA: GPU Optimization of Infomax-ICA EEG Analysis

In recent years, Independent Component Analysis (ICA) has become a standard to identify relevant dimensions of the data in neuroscience. ICA is a very reliable method to analyze data but it is, computationally, very costly. The use of ICA for online analysis of the data, used in brain computing interfaces, results are almost completely prohibitive. We show an increase with almost no cost (a rapid video card) of speed of ICA by about 25 fold. The EEG data, which is a repetition of many independent signals in multiple channels, is very suitable for processing using the vector processors included in the graphical units. We profiled the implementation of this algorithm and detected two main types of operations responsible of the processing bottleneck and taking almost 80% of computing time: vector-matrix and matrix-matrix multiplications. By replacing function calls to basic linear algebra functions to the standard CUBLAS routines provided by GPU manufacturers, it does not increase performance due to CUDA kernel launch overhead. Instead, we developed a GPU-based solution that, comparing with the original BLAS and CUBLAS versions, obtains a 25x increase of performance for the ICA calculation.

Measurement and Visualization of Dynamics of Piezoelectric Microcantilever

Methods for measuring the resonant frequencies and visualizing the motion of the Pb(Zr0.5Ti0.5)O3 microcantilever are investigated. Considering the two-segment structure of the microcantilever, a self-exiting self-sensing method is proposed to obtain the fundamental resonant frequency. An optical system consisting of light microscope, CCD camera and video card is established to visualize the first two vibration mode shapes. The theoretical, measured and visualized first resonance of one micocantilever is 17.28 kHz, 17 kHz and 17.8 kHz, respectively. A theoretical second resonance of 84.16 kHz is seen at 71.9 kHz. The proposed method is valid for measuring and visualizing low resonances of active micro structure.

Future of Thermal Management of High End Video Cards in ATX and BTX

The present study aims to examine the future of the thermal management of video cards in PC systems. The primary focus of this investigation is on high end video card applications in both the ATX and BTX platforms. In this article, the thermo-mechanical architecture of these platforms will be revealed along with a discussion of the power density road map of both the graphics processing unit (GPU) and video processor unit (VPU) expected to be in use until the year 2010. Recommendations of how to extend the air cooling limit for video cards in either the ATX or BTX platforms will be part of the conclusions of this study.