• Home |
  • Research & Development

Research & Development

Biomedical Sector

Animal Age, Dimension & Weight Prediction Solutions:

Duration: 1-year
Technologies: Python, OpenCV
  • Measurement of height, length & width.
  • Age & weight predictions.
  • Assess the health of the animal using High-definition cameras powered by AI solutions.

The goal of the project is to develop software for automatic calculation of animal weight based on the image of the animal with required accuracy not higher than 3%. The calculation of weight was based on an algorithm of a reconstruction of a 3D model of the animal-based on a set of images and estimating the weight using specific animal parameters such as length of the back, size of the torso, and others.

Sports

Video solution athletics platform:

Duration: 1-year
Technologies: Python, OpenCV
  • Saves time, storage space, Improves work accuracy
  • Removes unnecessary or idle content

The goal of the project is to analyze the video of a cricket game for breaking matches into shorter videos: one video per point. It was required to remove those parts of the match where the players did not play (the players rest, the gap between the points, etc.); that allowed game statisticians to make further revisions of the game much faster because all idle periods of the game were removed and the total length (as soon as file size) was much shorter. The logic of breaking video has been developed based on the analysis of game events that were detected in the video, position, speed, and posture of players, ball movement and location, and other parameters. CV algorithms were used: optic flow, background subtraction, HoG detector, pose detection, and others

Agriculture & Farming:

Basic image processing and image alignment:

Duration: 6 months
Technologies: С++, OpenCV, Qt, Pattern correlations
  • Helps in analyzing the growth of fruits or vegetables
  • Estimate the length & width of each row

The goal of the project is to analyze images of grape fields from drones, to find and detect grape rows, estimate the start and endpoints, length, and width of each row. Additionally, the application supports additional image processing functions such as colors, brightness, and contrast manipulation, drawing primitives (lines, polygons), zoom, saving results as to shapefiles, and supporting geolocation information.

Information Technology:

Looking for correlations and making predictions in e-commerce:

Duration: 5 months
Technologies: Python, Scikit-learn
  • Better planning of business & strategy
  • Understanding the customer interests & behavior
  • Increases revenue of a business

An implemented system that ranks customers making projections critical for business actions like upgrading from free to premium accounts, amounts of payments for premium accounts, and how likely that person will stop using the service. The system is trained on a database of user account information and user series of actions made at the service. After the new user data is provided to the system, a ranking based on projected future actions is produced. We used the following techniques a) feature engineering for machine learning; b) random-forest/decision trees models; c) bag of words like a model for user actions.

Natural Language Processing:

Looking for correlations and making predictions in e-commerce:

Duration: 1 year
Technologies: Python
  • Accurately capture the meaning and themes in text collections, and apply advanced analytics to text, like optimization and forecasting.
  •  Identifying the mood or subjective opinions within large amounts of text, including average sentiment and opinion mining.
  • Transforming voice commands into written text, and vice versa.
  • Automatically generating synopses of large bodies of text.
  • Automatic translation of text or speech from one language to another.

Developed an application that analyzes a database of CVs and vacancies and tries to find the closest matching pairs of CV and vacancy. The pair is detected using analysis of CV and vacancy text, detecting a tone of the text, matching criteria in vacancy, and finding corresponding criteria in CV text.

ChatBot:

Human-like chatbot in National languages:

Duration: 6 months
Technologies: Python, JavaScript
  • Provides 24/7 customer service
  • Serving customers in a better & effective way.
  • Capable of providing faster solutions
  • Analytics for types of questions asked

We Implemented a chatbot that replies to the question in a human-like format. Replies are based on the knowledge base of questions and answers in the National language. Additionally, the usage of the Q&A knowledgebase engine uses some «improving» techniques to provide better results, such as:

  • topic extraction for answers in Q&A knowledgebase;
  • mistyping’s corrections of users’ messages;
  • synonyms usage/replacement;
  • replies rating;

Internal R&D

NLP Chatbot Project:

Duration: 2 years
Technologies: Java, Python

For the last 2 years, we’ve been working on a chatbot technology recently becoming a trend hype due to the interest from big corporations like Facebook. We conducted experiments to find the best way to implementation of chatbot. We focused on using open-source tools written in popular languages (such as Python) mixed with our experience and perception of AI. The main goal of NLP is to make a computer-based system that will be able to communicate with a human using his/her native language, such as English, Russian, or any other. To achieve this goal several tasks need to be accomplished. Roughly these tasks can be grouped in three steps: text decomposition (information extraction), reasoning, and text generation. This is referred to as the NLP pipeline and usually includes the following subtasks (note that subtasks 1-3 correspond to the text decomposition phase):

Conversion of information that contains natural language to machine-readable text.

  1. Morphological and lexical analysis, parsing.
  2. Semantic analysis [2, 3], information extraction.
  3. Knowledge representation, reasoning.
  4. Natural language generation, similar to steps 1-3 but in backward order.

Chatbot

CMS using ChatScript:

Duration: 3 months
Technologies: C++, Python, Google Docs API

The goal of the project is to build a content management system that bases on Google Spreadsheets and ChatScript. CMS is designed and built, so it can be connected to Field book and Google Sheets with client chatbot app to spin up new chatbots on demand. Also, edits can be done in Google Sheets or Field book and then pushed to production. This is a solution for writers to create content directly into a CMS, organized around the ChatScript syntax. The system processes these images as learning samples to build a database and corresponding identification model. Based on the built model system tries to identify an author for any custom signature.

Safety and security

People Collapse/Fall detection:

Human fall detection is an area that has been a subject of research and development interest recently due to technological advancements in Artificial Intelligence and Machine Learning. The security and safety of elderly people have become an issue of concern. One specific health risk factor for the elderly is the increased likelihood of falls and the resulting bone fractures to various body parts. The goal is to detect the person collapse using HD cameras

Key features:
  • Helps to save the life of a person by hospitalizing ASAP
  • Maintains safety & security of a person
Border Security solution:

The rise of terrorist threats, smuggling, criminal activity, illegal immigration demands a robust security solution. The goal of the project is to develop a vision-based security solution that can safeguard national borders. Our solution includes:

  1. Detection of humans, abandoned objects, animals, vehicles near the border
  2. Detection of activity like crawling, digging, throwing
  3. Detection of UAVs, Weapons near the border
Key features:
  • Provides security to the national borders
  • Eliminates national threat by detecting unlawful activities
Fire and Alarm solution:

We deeply believe that the earliest and accurate fire detection plays a very critical role. It helps to evacuate the people in the building ASAP by alerting the control room. The goal of the project is to detect the fire in the premises /near the premises.

Key features:
  • Protects the life of people inside premises
  • Generates alert to control room for immediate evacuation
  • Activates alarm and water shower to stop the fire
People Safety & Security

The goal of the project is to detect a human with/without a PPE kit, Gloves, Mask, Face shield, boots, Safety Goggles, Helmet, and other safety gear. Which helps to protect the safety & security of an employee in a factory or a hazardous working environment.

Key features:
  • Detection of personnel and his safety gear.
  • The control room is alerted in case of safety & security violations.

Analytics

Automatic Number Plate Recognition (ALPR):

With increasing vehicles, it is very important to maintain the safety of the people traveling on the roads the usage ranging from monitoring traffic activities such as red-light adherence, speed enforcement, detection of theft, electronic toll collection to various other traffic enforcement applications, ANPR is technology-driven by how well it ensures the continued safety of all road users. Our ANPR system is based on artificial intelligence, providing a robust and ready-to-integrate system, capturing diverse types of license plates.

Key features:
  • Assists the enforcement agency as well as safeguards citizens
  • Alert is generated during violation of rules
  • Reduces road accidents
Track & Trace:

The goal of the project is to track a person by the color of apparel. Our system helps trace the person. User can flag suspicious persons and take necessary action

Key features:
  • Enhances the safety of premises
  • Helps in tracking the suspicious person
Crowd Mapping

Crowd mapping can be used to track fires, diseases like COVID-19, flood, pollution, crime, political violence, earthquake, and bring a level of transparency to fast-moving events that are difficult for conventional media to adequately cover.

Key features:
  • Provides better insights
  • Helps to curb the spread of diseases like COVID-19