The RACOON project (Coherent dual-comb lidar adapted to the marine environment) is part of the development of new underwater imaging techniques, both for military applications (threat detection, intrusion, monitoring of sensitive sites, etc.) and civilian (resource detection, submerged works monitoring, navigation safety and wreck inspection...). This project aims to validate the concept of a coherent underwater lidar, develop a prototype, then test it and characterize its performances under realistic conditions of underwater environment. RACOON therefore proposes to go beyond the state of the art of underwater lidar, which is based so far exclusively on an incoherent approach, that is to say that the signal detected is the light intensity. On the contrary, the coherent approach, which relies on the detection of the electric field, offers a priori important advantages. Indeed, in this approach, the signal varies as the inverse of the distance to the target (against the inverse square in the incoherent approach). In addition, the detection of the field makes possible the filtering of the photons diffused by the particles suspended in the sea water, and therefore must significantly increase the signal-to-noise ratio. Finally, coherent detection offers the possibility of measuring the movements or speeds of the target. First, we will demonstrate the relevance of the coherent approach on a 532 nm laboratory test system, which will allow a simple direct comparison of coherent and in coherent configurations. This test system will also be characterized during the project, in the seawater tank DEXMES of the Laboratoire Géo-Océan, in Brest, which allows to have controlled marine environments (turbidity, flow velocity). Then we will develop a coherent lidar prototype in the blue-green spectrum. This prototype is based on an architecture well mastered by the FOTON Institute, a double loop with frequency shift (or bi-directional loop). Initially, a 1550 nm loop will be made on the basis of the experience acquired by the laboratory on previous projects (ANR COCOA, ANR Astrid MECHOUI). A tripling frequency stage will reach the blue-green region (517 nm), suitable for underwater propagation. This prototype, which uses the principle of dual-comb and multi-heterodyne detection, must offer a sub-centimeter resolution. It will be tested first in the DEXMES tank, then, at the end of the project, in the IFREMER instrumented channel (50 m) which will characterize the performance of the prototype under realistic conditions, in terms of scope, resolution, signal to noise, and the ability to measure movements and speeds. In parallel, a technology watch task will be conducted on the sources and components in the blue-green spectrum. For the moment, the performances of these do not allow the realization of a dual-comb system directly in this spectral range, but we anticipate in the long run this possibility, which will greatly simplify future coherent underwater lidar systems. The RACOON project is a 36-month, single-partner project led by two teams from the FOTON Institute, in Rennes and Lannion. It will draw on the experience and know-how of external collaborators (). The work will be carried out by researchers, faculty members, and technical staff of the FOTON Institute, with the help of a postdoctoral researcher recruited for theis project.