DIMAP (DIMAP (Metadata format for SPOT / Pleiades remote sensing imagery))
Sep 24,2026

GISBox is a one-stop 3D GIS data editing, conversion and publishing platform that supports editing in multiple GIS formats such as OSGB/GEOTIFF/RVT, converting to 3DTiles/Terrain and publishing.

Introduction

DIMAP stands for Digital Image Map. It is an open remote sensing image distribution data format specification jointly introduced by the French company Spot Image, Satellus, and the French Space Agency CNES. The early version 1.1 was primarily used for SPOT 5 satellite data distribution. Subsequently, the iterated DIMAP v2 version was adapted for high-resolution satellites such as Pléiades. It can integrate GeoTIFF or JPEG2000 image files and comes with rich metadata including satellite orbit information, imaging parameters, Rational Polynomial Coefficients (RPC), and coordinate systems. It supports high-precision image geometric correction with few ground control points and can be directly read and parsed in mainstream remote sensing processing software such as ENVI and GDAL, greatly reducing the operational threshold for remote sensing image geometric processing.

File Structure

DIMAP is a standard format adapted for SPOT and Pléiades remote sensing imagery. Its file structure can be divided into the following core parts:

Main header file: An XML-format file with the .dim extension. It is the core entry point of the entire dataset and completely stores all metadata of the product, including key content such as imaging time, sensor parameters, radiometric calibration coefficients, Rational Polynomial Coefficients (RPC), and coordinate systems. Mainstream remote sensing software can automatically identify and load the corresponding imagery and accompanying parameters through this file.
Same-named data directory: A folder with the same name as the main header file and the .data extension. It internally stores the image files corresponding to each band and supports integration of ENVI-compatible formats, GeoTIFF, or JPEG2000 image data. The JPEG2000 format encodes imagery into independent 2048×2048-pixel tiles as the basic decompression unit.
Auxiliary subdirectories: Some versions include a tie_point_grids subdirectory that stores control point grid files accompanying the imagery. They also contain quality masks and cloud cover masks for subsequent image quality screening and processing.
Accompanying preview files: An additional KMZ-format preview file is included, which can be quickly loaded in Google Earth to view the imagery's coverage extent and basic metadata information, allowing users to quickly verify the product's spatial position and basic attributes.

Pros

Extremely high metadata integration: Comes with complete RPC rational polynomial coefficients, radiometric calibration coefficients, imaging time, sensor parameters, and other information, directly supporting high-precision orthorectification and greatly lowering the threshold for geometric processing.
Strong compatibility: Supports integration of both GeoTIFF and JPEG2000 image formats and can be directly recognized and loaded in mainstream remote sensing software such as ENVI and GDAL. It also includes KMZ preview files for quickly viewing imagery coverage in Google Earth.
Comprehensive accompanying information: Comes with quality masks and cloud cover masks. File naming rules are intuitive, allowing information such as sensor, imaging date, processing level, and band combination to be obtained directly from file names, facilitating batch data management.
Supports fast partial reading: JPEG2000 encoding uses a 2048×2048-pixel tile structure, making it suitable for fast browsing and processing of local regions in large, high-resolution imagery.
Open standard: Defined by France's CNES, it is an open format specification adopted by multiple generations of satellites including SPOT 5 and Pléiades 1A/1B, with a mature ecosystem.

Cons

Relatively low random access efficiency: If JPEG2000 encoding is used, reading even a single pixel requires decoding a complete 2048×2048 tile, making small-area pixel extraction time-consuming.
Vendor-customized format with limited extensibility: As a satellite vendor-customized distribution format, some niche versions lack universal standards for extended metadata fields, making it prone to incomplete parameter parsing when read by non-original tools.
Many files and inconvenient management: A complete dataset consists of multiple dispersed files including the .dim header file, .data directory, KMZ, and quality masks. Cross-device transfer and long-term archiving are prone to issues such as missing files and mismatches between metadata and imagery.
Limited compression and bit depth support: Some older versions do not support lossless compression of 16-bit imagery, resulting in relatively large storage footprints for large, high-resolution imagery. Additionally, JPEG2000 lossy compression may cause detail loss in high dynamic range scenes (such as snow/ice and shadow areas).

Application Scenario

The DIMAP format is primarily used for data distribution and production processing of high-resolution commercial satellite imagery such as SPOT 5, SPOT 6/7, and Pléiades 1A/1B. It widely serves scenarios requiring sub-meter spatial resolution, including fine-scale urban surveying and mapping, disaster emergency assessment, and crop monitoring. At the same time, because it comes with complete RPC parameters and radiometric calibration coefficients, it is also one of the most commonly used input formats in orthorectification and batch geometric processing workflows for remote sensing imagery. It can be directly parsed in mainstream software such as ENVI and GDAL, making it a standard intermediate carrier for commercial high-resolution remote sensing data from acquisition to application.