Who we are. What we do.

ABTO Software started the science journey 10 years ago as a partner of Lausanne University in Switzerland. One of our first projects was to develop smart software for horizontal drilling. The solution was recognised as the world’s most advanced steering tool for laying pipes and the first guidance system not affected by magnetic interference.

Scientists and innovative decision makers across all industries come to us with their ambitious ideas. We build and train intelligent applications that help businesses improve safety of people’s homes, fight fraud, reduce number of traffic incidents, or simply bring color to your black and white photos.

LEARN MORE…

Machine Learning in Real-World Applications by Industry

Machine learning can provide insight for all businesses and industries, tackling challenges of text analysis and natural language processing, object recognition.

Banking and Finance

  • Fraud prevention
  • Risk Management
  • Investment and trading opportunities
  • Money laundering

Government and Security

  • Identity theft prevention
  • Face recognition
  • Image classification
  • Speech recognition

Healthcare

  • Alerts and diagnostics in medical imaging
  • Real-time treatment recommendations
  • Drug Discovery
  • Robotic Surgery

Marketing and Sales

  • Personalized content recommendations
  • Customer sentiment analysis, or opinion mining
  • Demand forecasting
  • Customer behaviour prediction

Energy & Utilities

  • New energy sources
  • Renewable Energy
  • Power and other utilities usage analytics
  • Energy demand and supply optimization

Transportation

  • Traffic prediction and congestion management
  • Route planning
  • License plates recognition
  • Vehicles recognition/detection

Projects

We gather relevant datasets and develop intelligent applications that can read, learn and advise like humans do. Learn what problems ABTO Software has helped solve for the clients today, and help us find your solution tomorrow.

Customer:

Tech company, Israel

Project challenges:

  • Camera software works in real-time environment under any conditions: illumination, weather, etc
  • Detect vehicle direction
  • High software performance, since in most cases the vehicle is a moving object
  • Powerful synergy with the client’s suite of vehicle recognition products
  • Poor input file resolution, usually because the plate is too far away, or blurry images, particularly motion blur
  • Simultaneous processing of multiple video streams from CCTV cameras on different devices

Technologies and Instruments:

  • The back-end development tech stack: PHP, Laravel, Node.js, Amazon Web Services: S3, RDS, SQS, SNS, Database: MySQL.
  • Web UI tech stack: AngularJS, HTML, CSS.
  • Data Modeling and Architecture: machine learning and image/video processing algorithms, Haar Cascades.

Customer:

R&D project

Key features:

  • 75% precision rate
  • identification of food items from consumer camera-enabled device
  • real-time calorie counting
  • simple, lean and intuitive interface
  • possibility to add and edit food composition

Project challenges:

  • to train an accurate classification model based on an extensive food images database
  • to build a robust system with multi-view image variations
  • to implement computer vision technologies, clustering and machine learning for Android and iPhone devices

Technologies and Instruments:

  • Android, iOS, image processing, machine learning, feed-forward artificial neural networks, image segmentation and clustering (k-means)
  • MATLAB, C++, Python, OpenCV, Qt

Customer:

Expert consulting company, US

Benefits:

  • Increased Conversions. Being a machine-learning based conversational dialog engine, the chatbot can communicate with any number of people at the same time, giving instant responses to the questions users present.
  • Higher engagement. Customers are twice more likely to interact with a website when you reach out with a proactive chat

Project challenges:

  • Optimization of the dialogues and reactions to the users’ messages for meaningful conversations outcomes
  • Ensuring basic language understanding and specific scenarios
  • Personalization

Technologies and Instruments:

  • AAI, ChatScript, flask, gunicorn, machine learning algorithms, nginx, Python, scikit-learn

Machine Learning Tools and Technologies

Machine learning algorithms, methods and tools we use to develop smart software that can recognize objects, features, and patterns within images, video, and written content generating highly intelligent and insightful data.

Supervised learning algorithms

  • Artificial neural networks
  • Decision tree learning
  • Learning classifier systems
  • Nearest neighbor algorithm
  • Support vector machines
  • Random forests
  • Ensembles of classifiers
  • Bayesian statistics
  • Naive bayes classifier
  • Convolutional neural networks
SEE MORE

Unsupervised learning

  • Clustering
    • K-means
    • Mixture models
    • Hierarchical clustering
  • Anomaly detection
  • Approaches for learning latent variable models such as
    • Expectation–maximization algorithm (EM)
    • Method of moments
    • Blind signal separation techniques, e.g.
      • Principal component analysis
      • Independent component analysis
      • Non-negative matrix factorization
      • Singular value decomposition
SEE MORE

Tools

Investigation:

  • MATLAB
  • R
  • Python

Implementation:

  • C++
  • OpenCV

Contact Us

If you see a partnership opportunity or have a challenging project

Get in touch with us!