GBT Tokenize Is Developing Real Time Object Detection Algorithms And Techniques For Kirlian Research
SAN DIEGO, Feb. 23, 2021 (GLOBE NEWSWIRE) -- GBT Technologies Inc. (OTC
PINK: GTCH) ("GBT” or the “Company”), together with GBT Tokenize Corp (“GBT/Tokenize”) is developing real time algorithms and techniques for Kirlian research. A Kirilian image includes
vast amount of graphical data. GBT/Tokenize research is aiming to develop a system and method to analyze this data as a potential of health-related information source. These algorithms and methods
are pattern detection and recognition, based on unique principles. An image’s objects are categorized according to their physical characteristics like shape, color, texture, and more. A machine
learning based flow is targeted to operate as an object’s classifier and analytics processor. We believe that the object detection and analytics method can be used for processing Kirlian images for
their characteristics as another stage of GBT/Tokenize’s Kirlian Electrophotography research. Kirlian photography method introduces a series of techniques that are based on the phenomenon known as
electrical coronal discharge. This technique produces an object’s energy related images with a colorful representation called aura. When performed on human organs, although not scientifically
proven, some believe that these images can be interpreted to analyze health conditions. GBT/Tokenize is researching the development of imaging related techniques to further investigate the data
generated from these images for possible health related conclusions. The research is not medical but technical and targeted to conclude a real time analysis of Kirilian images that may be related
to human’s health conditions.
The goal of our research is to enable a machine learning algorithm to decide as to whether an image’s object is of interest or not, pointing a possible health related conclusion. A Kirlian image contains a huge amount of data. The detection and analytics of a Kirlian image requires major computational capabilities in a real time operation. A neural network analysis is done to ensure a reliable object’s classification and to train for the detection of objects of interest within a Kirlian image. The approach performs an image based color-based pre-processing, to reach a conclusion about certain pattern and color presented in the image. The goal is to reach a real-time Kirilian image processing with the use of deep learning algorithms and supporting computational hardware resources, achieving advanced imaging conclusions that may provide health related information.