Artificial Intelligence, Computer vision and Object Recognition

A specific platform for the insurance sector.

An image analysis engine system based on recurrent artificial neural networks that carries out anti-fraud checks, extrapolates and contextualizes information and quantifies the damage caused by the car accident.

What is Horus?

A platform of intelligent engines consisting of recurrent artificial neural networks, specific for the insurance sector.

Thanks to complex processes based on Artificial Intelligence, Computer vision and Object Recognition techniques, Horus deals with identifying, recognizing, interpreting, classifying, checking and making the data contained in the analyzed images usable.

Horus has three souls, born to fulfill three particular tasks


extract information from documents of any structure and complexity into data ready to be used in the enterprise, in digital or electronic format

Learn more


analyze images and documents to detect anomalies, graphic manipulations, visual inconsistencies and guarantee their uniqueness on db and online

Learn more


recognize the damaged parts of the photographed accident vehicle and make an automatic estimate of the repair costs in real time

Learn more

How Horus works

The main purpose of HORUS is to identify the visual information contained in the image and transform it into manageable digital information for the most diverse uses.

A smartphone is enough to send the document to HORUS and submit it to an in-depth automatic analysis.

Using self-learning algorithms, data mining, pattern recognition, signal processing and with the support of the most advanced hardware technologies, Horus performs a series of activities on the images starting from simple ones to get to increasingly complex processing, imitating the modes of the human brain. .

Streamline and speed up thanks to Artificial Intelligence

Automation through artificial intelligence brings with it a significant opportunity to optimize decision-making processes and reduce management costs and is therefore destined to permeate all aspects of the insurance business, allowing employees and organizations to focus more on business. of greater value as well as on works that require more imagination and creativity.

The assignment of repetitive and low-added-value activities to “intelligent” tools and technologies not only leads to an improvement in process efficiency, but above all ensures compliance with increasingly stringent regulatory requirements for verifiability, security, data quality and operational resilience .

ICG offers customized Artificial Intelligence and Machine Learning solutions for Insurance Players

ICG supports companies in exploiting the potential of Artificial Intelligence to implement sector-specific solutions. Thanks to the skills consolidated during the implementation of numerous PoCs with latest generation technologies, it can be considered at the forefront of these issues that will inevitably mark the near future of the insurance world.

[grid_plus name=”AI_products”]


intelligent digitization of documents and automation of document processes

An innovative platform capable of interpreting and processing data from all types of documents, including paper (scanned or photographed) or digital, structured and unstructured documentation formats, images, email attachments and message bodies.

Thanks to the intelligent digitization of documents and the automation of document processes , it transforms documents into value for the company, capturing and validating information in any format at the point of use.

Learn more


Fighting insurance fraud related to false documents

An Artificial Intelligence platform that deals with analyzing images and documents of any structure and complexity to recognize graphic manipulations, visual inconsistencies, verify the uniqueness of the image within the DB, check that the image is not present on the internet. To combat insurance fraud related to false documents.

Learn more



Artificial Intelligence, Computer vision and Object Recognition to recognize the damaged parts of the photographed vehicle, distinguish between heavy and light damage and make an initial estimate of the repair both in terms of costs and times, channeling complex and mass claims in a totally automatic way .

Learn more

The learning models of neural networks

In order for this process to be efficient, it is necessary to “train” the neural networks, that is to make them learn how to behave when an engineering problem has to be solved, such as the recognition of a human being from the analysis of images (through for example facial recognition technology).

A neural network actually looks like an “adaptive” system capable of modifying its structure (nodes and interconnections) based on both external data and internal information that connect and pass through the neural network during the learning phase and reasoning.

The revolution brought about by Deep Learning is evident on many fronts, so much so that starting from 2015 algorithms that exploit deep neural networks have exceeded the capabilities of man in simple tasks such as image recognition and transcription of audio into text.

The merits of Deep Learning are to be found in the ability to obtain ever better results by increasing the complexity of the neural network or by adding unstructured input data to the model. From the point of view of performance, the possibility of exploiting GPUs for massive computations has also favored the introduction of more complicated algorithms in production environments.

Insurtech and surroundings


What are the applications in the insurance sector?

Find out more

Contact us to find out how we can help your business.

Learn more