A5060623456- Cell Image Analysis Techniques
- Advanced Image and Video Retrieval Techniques
- Advanced Electron Microscopy Techniques and Applications
- Advanced Vision and Imaging
- Computer Graphics and Visualization Techniques
- Robotics and Sensor-Based Localization
- Medical Image Segmentation Techniques
- Advanced Fluorescence Microscopy Techniques
- Industrial Vision Systems and Defect Detection
- Image Processing Techniques and Applications
- 3D Shape Modeling and Analysis
- Machine Learning and Data Classification
- Brain Tumor Detection and Classification
- Molecular Biology Techniques and Applications
- Image Processing and 3D Reconstruction
Singidunum University
2020-2024
Recent advances in computing power triggered the use of artificial intelligence image analysis life sciences. To train these algorithms, a large enough set certified labeled data is required. The trained neural network then capable producing accurate instance segmentation results that will need to be re-assembled into original dataset: entire process requires substantial expertise and time achieve quantifiable results. speed-up process, from cell organelle detection quantification across...
Identification and quantitative segmentation of individual blood vessels in mice visualized with preclinical imaging techniques is a tedious, manual or semiautomated task that can require weeks reviewing hundreds levels data sets. Preclinical imaging, such as micro-magnetic resonance (μMRI) produce tomographic datasets murine vasculature across length scales organs, which outmost importance to study tumor progression, angiogenesis, vascular risk factors for diseases Alzheimer's. Training...
Recent advances in computing power triggered the use of Artificial Intelligence image analysis life sciences. To train these algorithms, a large enough set certified labelled data is required. The trained neural network then capable producing accurate instance segmentation results, that will need to be re-assembled into original dataset: entire process requires substantial expertise and time achieve quantifiable results. speed-up process, from cell organelle detection quantification across...
This paper proposes a solution for distance estimation using stereo images.The is convolutional neural network that takes two images as an input, and outputs the estimate, without need prior camera calibration or disparity map calculation.The dataset used training consists of generated from artificially constructed 3D scene.The algorithm was stochastic gradient descent.Evaluation conducted on separate dataset.Mean absolute error after evaluation 1.59 m, while median value 1.2 m.These results...
The rendering of a large number images is demanding task, usually delegated to remote farms.This necessitates the creation software for management these tasks.Several commercial and noncommercial solutions can be used this task.However, they are not specialized rendering, thus require either additional configuration expansion, or familiarity with machines manual setup tasks.In paper we present solution alleviate issue.The consists two components: Microraptor GUI client.Microraptor...
Surface reconstruction from low quality point clouds represents a common problem in most standard algorithms created for this purpose.Point acquired using specialized devices, such as 3D scanners, or outputs structure motion are usually flawed that they contain significant amount of noise and outliers, making the surface process difficult, resulting estimation.The reconstructed mesh is directly proportional to cloud itself.This paper proposes workflow creating surfaces unstructured...