Could Belgium’s IDDEA MEGA Solution Transform Battlefield Equipment Identification?.

Belgium-based IDDEA is developing a mobile app that identifies over 1,700 military systems from photos or video. The capability could accelerate battlefield decision-making and reduce misidentification risks for U.S. and allied forces.

The Military Equipment Guide App, or MEGA, uses image recognition to classify armored vehicles, artillery, and support systems in near real time via a smartphone interface. Demonstrated at BEDEX 2026, the system is now moving toward fully offline functionality through an industrial partnership, a shift aimed at enabling use in denied or degraded communications environments. IDDEA positions the tool as a rapid reference aid for frontline troops, intelligence units, and training pipelines.

Related topic: Belgium’s IDDEA Develops Offline App Giving Soldiers Instant Access to Critical Military Data

Belgium’s MEGA app uses AI to identify 1,700 military systems from smartphone images, aiming to improve battlefield awareness. (Picture source: IDDEA)

A mobile application developed in Belgium could alter battlefield identification practices by providing soldiers with rapid access to a database of military vehicles and equipment. The Andenne-based company IDDEA is developing the Military Equipment Guide App (MEGA), designed to recognize more than 1,700 systems from a simple photo or video captured via smartphone, with the aim of simplifying analysis in combat conditions. The initiative gains visibility in early 2026, notably after a presentation at the Brussels Defense Exhibition Day (BEDEX), where the company enters into an industrial partnership intended to transition the system toward fully offline operation.

The project is rooted in operational experience. IDDEA’s founder, Alain Servaes, a former military professional, identified in 2019 a persistent gap in identification tools used by some armed forces. At that time, basic supports such as card decks were still used to recognize adversary equipment. This limitation leads to the development of a digital solution capable of centralizing visual references and technical data in a format directly usable in the field.

As reported by L’Avenir, the application relies on a database built over several years, consisting of hundreds of thousands of images collected during international exhibitions, field observations, and open sources. Each entry is annotated to enable precise identification, including vehicle type, main characteristics, and known capabilities. The architecture is based on multi-angle and multi-condition indexing, improving recognition when images are partial or degraded.

In its initial configuration, MEGA operates as a recognition software requiring a mobile network connection to access the database. This dependency represents a limitation in operational environments where emissions can be detected. The agreement signed at BEDEX marks a shift in development. A first step allows the database to be downloaded directly onto the device, enabling identification without connectivity. However, the objective extends further. IDDEA is working with its partner on a fully embedded solution combining both the recognition engine and the database locally, without reliance on external infrastructure.

This approach changes the nature of the system. It is no longer limited to a mobile application but has evolved into an embedded software that can be integrated into military vehicles or optronic systems. Image processing and identification are performed locally, which requires optimization of algorithms and storage. Early indications suggest that recognition can be achieved within a few seconds, depending on image quality, despite the computational constraints associated with embedded hardware.

The current scope focuses primarily on land systems, including armored vehicles, infantry fighting vehicles, and logistics assets, while expanding toward other domains. Unmanned systems, naval units, and threats such as mines are progressively being added. The identification of loitering munitions such as the Iranian-designed Shahed series reflects this expansion. These systems typically have a range of approximately 1,000 to 2,000 kilometers and carry a warhead of around 30 to 50 kilograms, making early detection relevant for both air-defense units and ground forces.

From an operational perspective, this development introduces practical implications. When integrated into an armored vehicle or an observation post, the system could analyze real-time feeds from optical sensors and provide near-immediate identification. In high-intensity environments where hundreds of armored vehicle variants may be present, as observed in Ukraine, such a capability reduces uncertainty and supports faster decision-making. It may also help limit misidentification, which remains a recurring issue in complex engagements involving multiple actors.

The absence of connectivity further enhances discretion. By eliminating emissions associated with network use, the system reduces electromagnetic signature, which is relevant against adversaries equipped with detection and electronic-warfare capabilities. Embedded integration also enables continuous use regardless of network availability.

Beyond the tactical level, this development reflects a broader trend toward distributing analytical capabilities to lower echelons. The transition from a connected application to a fully autonomous embedded solution illustrates an effort to design systems adapted to contested environments. If adopted at scale, such technologies could influence identification practices and information-sharing approaches, particularly in multinational contexts where interoperability and data standardization remain ongoing challenges.