A5037723791- Autonomous Vehicle Technology and Safety
- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Remote Sensing and LiDAR Applications
- Fire Detection and Safety Systems
- Advanced Vision and Imaging
- Currency Recognition and Detection
- Blind Source Separation Techniques
- Image Processing and 3D Reconstruction
- Single-cell and spatial transcriptomics
- Vehicle Dynamics and Control Systems
- EEG and Brain-Computer Interfaces
- 3D Surveying and Cultural Heritage
- Automated Road and Building Extraction
- ECG Monitoring and Analysis
- Digital Media Forensic Detection
- Cell Image Analysis Techniques
- Advanced Proteomics Techniques and Applications
Innopolis University
2018-2022
Abstract While spatial proteomics by fluorescence imaging has quickly become an essential discovery tool for researchers, fast and scalable methods to classify embed single-cell protein distributions in such images are lacking. Here, we present the design analysis of results from competition Human Protein Atlas – Single-Cell Classification hosted on Kaggle platform. This represents a crowd-sourced develop machine learning models trained limited annotations label patterns fluorescent images....
Automatic illegal building detection from satellite imagery is a specific and important problem for both research community government agencies, which has not been sufficiently investigated since it combines the challenge of automatic remote sensing data interpretation verification with cadastral map. Recovery footprints images very complicated process because areas their surroundings are represented various color intensities complex features. This paper proposes methodology that...
Self-driving vehicles contain a number of modules allowing them to autonomously navigate in uncertain environment. The robust, efficient, safe and accurate autonomous navigation are heavily depend on parameters perception module. In this paper, we consider module as combination object detection road segmentation submodules. As matter fact, all based Deep learning technique. It leads liability big training datasets provide the accuracy, efficiency robustness for self-driving car operating...
Human heartbeat can be measured using several different ways appropriately based on the patient condition which includes contact base such as by instruments and non-contact computer vision assisted techniques. Non-contact approached are getting popular due to those techniques capable of mitigating some limitations contact-based especially in clinical section. However, existing guided approaches not able prove high accurate result various reason property camera, illumination changes, skin...
Despite the significant advancements in object detection, there is still a number of detection tasks that are challenging and difficult. To eliminate them to push forward research this area technological "IceVision" Challenge was organized Russia 2019. The main objective develop new approaches for severe weather conditions focusing on winter time. obtained results supposed be implemented autonomous driving. This paper reports approach used by our team competition. based Cascade R-CNN model...
Localization is still one of the most challenging tasks in autonomous driving on city roads. Further development and improvement automatic functions vehicles urban conditions are not possible without overcoming problem significant degradation GNSS signal quality. The proposed approach to localization can provide information about vehicle position road different operational conditions. Desired stability quality achieved by using combination conventional computer vision, neural networks Kalman...
In this work we show our developed model for a self-driving car that solves road segmentation task and at the same time classifies whether vehicle is moving in or off lane. We demonstrate how scene classifier can be efficiently embedded into neural network by means of branching additional layers with small number trainable parameters. The was trained tested on own dataset good accuracy performance. Experiment external proves efficiency proposed approach shows reasonable results both...