Computer vision

Lingaro’s Research and Development teams have made computer vision a key field of focus since 2000. As a result, they are proficient in:

  • optical character recognition and simple pattern scanning technologies used in document identification systems, and
  • a wide range of the most important computer vision and artificial intelligence technologies based on algorithms related to boosting, support vector machine, locality-sensitive hashing, and discrete phase models, among others.

Applications span a wide variety of business use cases including people tracking, object detection, parking lot vacancy detection, 3D scanning, global content based image retrieval, and contextual document scanning, to name just a few.

Business-ready computer vision systems from Lingaro allow you to:


Optimize on-shelf visibility

Verify the location of products and ensure their optimal presentation, proper categorical placement, and proportional share of shelf space. PSE (Product Search Engine) is a patented multi-object multi-detection system build as part of a project financed by the European Union Regional Development Fund. The PSE’s novel approach to image analysis involves providing a graphical vote space presentation, generating a proposition, aggregating votes in two iterative passes, and verifying the proposition with two cascade filters containing all minor algorithms necessary for effective object detection. Its accuracy is exceptionally high in comparison to that of systems taking the classic clustering approach to analysis.

Check the visual coherence of products' packaging

Identify similar product packaging, check the similarity of one product’s packaging vs. that of others, and design packaging that is consistent across multiple products. NEOS is a content-based image retrieval system that can perform advanced similarity analysis of the colors, shapes, and textures in thousands of images.


Maximize retail operational performance

Ensure forecast accuracy, location marketing effectiveness, scheduling efficiency, low-hassle queuing, and good sales results for your retail location. PTBA (People Tacking and Behavior Analysis) monitors customer engagement and location during time spent in a store.

Keep your product visually competitive

Judge the visual symmetry of your product’s packaging and ensure that it is not overshadowed by that of other products on the same shelf. VSA (Visual Saliency Analysis) performs deep spectral analysis of a product image in many scales to represent observational distances, identifies its most important and visually appealing graphical elements such as its logo, and then then compares the elements to others within the same image as well as to those of nearby competing products.

Lingaro PRIP

Parallel Rasterization Image Processing

Lingaro’s PRIP is a unique software solution for rendering. Fast and efficient code offers order of magnitude speed improvements over single-threaded solutions. It is specifically optimized for high quality anti-aliased graphics reproducing the look of a printed page. Open source plugins for web browsers for image comparison are available here: https://github.com/LingaroRD/wpComparator

Request callback

Artificial Intelligence and Data Science Section Director

Błażej Chodarcewicz

Blazej has a strong, goal-oriented combination of technical and leadership experience gained over 15 years overseeing international teams — sometimes with as many as 200 members — and developing IT solutions based on a broad range of Enterprise Resource Planning, Customer Relationship Management, Business Process Management, Business Intelligence, Data Warehousing, and Web and Mobile application technologies. At Lingaro he focuses primarily on delivering Business Intelligence and Data Warehousing solutions that best fit customer needs in Consumer Solutions and Demand Planning initiatives. He holds an MSc in Computer Science from the University of Warsaw’s Faculty of Mathematics, Informatics, and Mechanics.