Circular No. 10/2015/TT-BTNMT stipulates technical requirements for producing high-resolution and ultra-high-resolution optical remote sensing images to be provided to users.

CIRCULAR

Technical regulations on the production of high-resolution and ultra-high-resolution optical remote sensing images for supply to users

_____________________________ 

Pursuant to Decree No. 21/2013/NĐ-CP dated March 4, 2013 of the Government on the functions, tasks, powers, and organizational structure of the Ministry of Natural Resources and Environment;

Considering the proposal of the Director of the National Remote Sensing Center, the Head of the Science and Technology Department, and the Head of the Legal Department;

The Minister of Natural Resources and Environment promulgates this Circular on technical regulations for the production of high-resolution and ultra-high-resolution optical remote sensing images for supply to users.

PART I

GENERAL PROVISIONS

Article 1. Scope of Regulation

This Circular stipulates the technical requirements for the production of high-resolution and ultra-high-resolution optical remote sensing images for supplying remote sensing images to users.

Article 2. Applicability

This Circular applies to state management agencies for remote sensing, organizations and individuals related to the activity of producing high-resolution and ultra-high-resolution optical remote sensing images.

Article 3. Explanation of Terms

In this Circular, the following terms are understood as follows:

1. Production of remote sensing images is the process of processing to create Level 1A, Level 2A, Level 3A, and Level 3B remote sensing image data from Level 0 remote sensing data.

2. Optical remote sensing images are types of images created by receiving visible light wavelengths (0.4 - 0.76 micrometers) through satellites using optical lenses.

3. Ground resolution (Ground Resolution) is the size of one pixel of an image projected onto the ground.

4. High-resolution remote sensing images are image data obtained with ground resolution less than or equal to 20 meters.

5. Ultra-high-resolution remote sensing images are image data obtained with ground resolution less than 1 meter.

6. Digital elevation model (Digital Elevation Model - DEM) is a numerical representation of continuous changes in surface terrain elevation. The digital elevation model is used in orthorectification to eliminate geometric distortions caused by terrain elevation differences.

7. Orthorectification is the correction of geometric distortions of images due to the imaging process, terrain elevation differences, and coordinate transformation from image pixel coordinates to Cartesian coordinates.

8. Enhancement of remote sensing image quality is the radiometric correction of images to improve their information content.

9. Level 0 remote sensing data is remote sensing data received and processed directly from satellite signals at the remote sensing image receiving station without undergoing any correction process.

10. Level 1A remote sensing images are remote sensing image products that have been corrected for the effects of Earth's curvature, receiver errors, and atmospheric noise from Level 0 remote sensing data.

11. Level 2A remote sensing images are remote sensing image products rectified to the national coordinate system VN-2000, international UTM plane coordinate system... using physical models, satellite orbit information.

12. Level 3A remote sensing images are remote sensing image products rectified to map projection coordinate systems and the national coordinate system VN-2000, international UTM plane coordinate system... using physical models, control points, and digital elevation models. Level 3A images are processed spectrally and enhanced in image quality by scene.

13. Level 3B remote sensing images are remote sensing images processed at the Level 3A stage, mosaicked and cut according to map tiles.

Article 4. Specialized Terminology

1. VN-2000: National Coordinate System VN-2000.

2. UTM: (Universal Transverse Mercator): International UTM Plane Coordinate System using transverse cylindrical projection.

3. GPS (Global Positioning System): Global Positioning System.

4. GIS (Geographic Information System): Geographic Information System.

5. Pan (Panchromatic): Full-color image.

6. XS (MultiSpectral): Multispectral image.

Article 5. Content of Remote Sensing Image Production Work

1. Preparation work: Collecting data and other technical documents related to the work.

2. Data entry: Entering remote sensing image data, map data, digital elevation models, and other relevant data into the remote sensing image processing system.

3. Producing Level 1A remote sensing images: Using collected data to produce Level 1A remote sensing images, then checking the quality of the product.

4. Producing Level 2A remote sensing images: Using collected data to produce Level 2A remote sensing images, then checking the quality of the product.

5. Producing Level 3A remote sensing images: Using collected data to produce Level 3A remote sensing images, then checking the quality of the product.

6. Producing Level 3B remote sensing images: Using collected data to produce Level 3B remote sensing images, then checking the quality of the product.

7. Packaging and submitting the product.

The process flow for producing remote sensing images is specified in Appendix 01 issued together with this Circular.

Chapter II

TECHNICAL REQUIREMENTS FOR REMOTE SENSING IMAGE PRODUCTION

Article 6. Data Collection

Data to be collected for remote sensing image production includes:

1. Remote sensing image data.

2. Map data.

3. Control point data for images.

4. Digital elevation model.

5. Project technical design and other related technical documents.

Article 7. Requirements for Remote Sensing Image Data

Remote sensing image data must ensure the technical parameters of remote sensing images and satellite orbits.

1. Level 0 remote sensing data must ensure quality during signal reception and processing, without signal errors or interruptions.

2. Level 0 remote sensing data must be preliminarily evaluated for cloud cover in the form of quick-view images. Cloud cover evaluation is divided into levels and marked with letters:

a) Level A: Remote sensing images with less than 10% cloud cover.

b) Level B: Remote sensing images with 10% to 25% cloud cover.

c) Level C: Remote sensing images with 25% or more cloud cover.

Selected remote sensing images for input data must have cloud cover suitable for user requirements.

Article 8. Requirements for Map Data

Maps used to determine control points for images must be national digital topographic maps at scales larger than the survey area of the remote sensing images to be established.

Article 9. Requirements for Control Points on Images

1. Selection criteria for control points on images:

Selected objects as control points on images must be clear on both remote sensing images and maps, not distorted due to map generalization or symbol size displacement, and must allow for height interpolation. The error in identifying and marking control points selected from the map must not exceed ±0.1 mm.

a) At intersections between objects or edges of linear objects (roads, dikes, dams, canals, streams) with angles of intersection between 30° and 150°;

b) At distinctive points of independent objects (bridges, culverts, corners of buildings);

c) Control points on images must be selected and marked on printed remote sensing images derived from original images.

2. Number and arrangement of control points on images:

a) Control points on images must control all scenes within the area. The minimum number of control points per scene is 12, evenly distributed in each remote sensing image scene. In the block of remote sensing images, there must be at least two control points located in the overlapping area between two scenes;

b) Each scene must have at least two check points, arranged between control points;

c) Coordinates of control points on images are determined on digital maps using information extraction tools from image processing software, digital maps, or GIS;

The layout design diagram for control points on images is specified in Appendix 02 issued together with this Circular.

Article 10. Requirements for Digital Elevation Models

The digital elevation model shall be used to rectify remote sensing images in cases where the positional displacement of image points caused by terrain height differences is ≥ 0.3 mm according to the scale of the remote sensing image map. In other cases, the average plane of the entire area shall be used for image rectification.

The permissible errors of the digital elevation models used for remote sensing image rectification are specified in Table 1 and Table 2 of Appendix 03 issued together with this Circular.

Chapter III

PROVISIONS ON DATA ENTRY

Article 11. Entry of Input Remote Sensing Image Data

1. Digital remote sensing image data must be converted to a suitable data format before being entered into the remote sensing image processing system.

2. Similar remote sensing film images must be scanned on film scanners with spectral resolution defined in units of 8-bit single-channel grayscale. The scanned data format must be compatible with the image processing system.

Article 12. Entry of Control Point Data for Remote Sensing Images

1. The error of control points transferred onto digital images compared to objects on the images must not exceed 0.5 pixels.

2. Control points must be assigned coordinates and heights based on the results of field control measurements or determined from larger-scale topographic maps.

3. Images, positions, and descriptions of control points must be stored in the database of the remote sensing image processing system.

Article 13. Entry of Map Data and Digital Elevation Models

Map data and digital elevation models must be entered into the remote sensing image processing system in accordance with the correct data formats.

Chapter IV

TECHNICAL PROVISIONS FOR PRODUCING REMOTE SENSING IMAGE 1A

Article 14. Entry of Level 0 Remote Sensing Data

1. Search for level 0 remote sensing data that meets cloud cover requirements and the date of image acquisition, presented as quick-look remote sensing images.

2. Check the storage location of level 0 remote sensing data stored in the storage system or on magnetic tapes.

3. Retrieve level 0 remote sensing data from the storage system or magnetic tapes.

4. Enter level 0 remote sensing data into the remote sensing image processing system to produce remote sensing image 1A.

Article 15. Production of Remote Sensing Image 1A

1. Set up processing configurations for technical parameters of remote sensing images.

2. Set up processing configurations for the orbit of remote sensing images.

3. Use software on the image processing system to produce remote sensing image 1A.

Article 16. Quality Inspection of Remote Sensing Image 1A

After production, remote sensing image 1A must undergo quality inspection:

1. Inspect cloud cover against set requirements.

2. Inspect spectral uniformity.

Chapter V

TECHNICAL PROVISIONS FOR PRODUCING REMOTE SENSING IMAGE 2A

Article 17. Entry of Input Data

1. Remote sensing image 1A, after passing quality inspection and meeting requirements, must be entered into the remote sensing image processing system as input data for producing remote sensing image 2A.

2. Enter the physical model of the remote sensing image.

Article 18. Production of Remote Sensing Image 2A

1. Calculate the physical model.

2. Select coordinate systems and projections that meet requirements.

3. Use software on the image processing system to adjust and create remote sensing image 2A. Pixel gray value interpolation is performed using the nearest neighbor method, bilinear method, or cubic interpolation method.

4. Process spectra using histograms for each spectral band.

5. Blend images: To enhance image quality for remote sensing image interpretation, mix full-color and multispectral processed images to create high-resolution color images.

6. Enhance image quality through filtering methods.

Article 19. Quality Inspection of 2A Remote Sensing Images

After production, 2A remote sensing images must undergo quality inspection as follows:

1. Coordinate system and projection zone inspection according to requirements.

2. Inspect spectral uniformity.

3. If 2A remote sensing images use image blending methods, accuracy between full-color and multispectral images must be checked, with positional errors not exceeding 01 pixel for scenes captured at the same time.

Chapter VI

TECHNICAL PROVISIONS FOR THE PRODUCTION OF 3A REMOTE SENSING IMAGES

Article 20. Input Data Entry

1. 1A remote sensing images that have passed quality inspection are entered into the remote sensing image processing system as input data for the production of 3A remote sensing images.

2. Enter the physical model of the remote sensing image.

3. Enter control points for remote sensing images.

4. Enter maps and digital elevation models.

Article 21. Thickening of Remote Sensing Image Blocks

1. The modeling of remote sensing images includes:

a) Entering remote sensing images: entering image data and accompanying auxiliary data (satellite information, satellite orbit, date of image capture, image position, camera information...) into the image processing system, along with control points and digital elevation models.

b) Thickening scene images: using satellite parameters during image capture, satellite orbit parameters, and image parameters to build physical models.

c) When connected into a block image (thickening of remote sensing image blocks), scenes, bands, and dual bands from satellites capturing at different times share a common distortion model. Use the block image model to align orthogonal scenes.

2. Selection criteria for band and block connection points:

a) Select band connection points, dual band connection points, and block connection points within the 10% overlap area of the band.

b) Connection points selected should be objects clearly visible on two scenes.

c) Connection points should be staggered, avoiding positions that form straight lines.

3. Checking thickening results:

a) Use existing GPS control points to calibrate scene image models, band models, or block models.

b) Calculate model errors; if errors exceed allowable limits, control points must be reselected. Positional errors of control points on the plane after error correction compared to the nearest GPS control point should not exceed 0.5 pixels, and in hidden areas, errors should not exceed 01 pixel.

Article 22. Geometric Correction of Remote Sensing Images

1. Remote sensing images are corrected individually per scene. The size of the pixel in the corrected image should not exceed 0.2mm on the map. Pixel grayscale values are interpolated using the nearest neighbor method, bilinear interpolation, or cubic interpolation.

2. Remote sensing image modeling is performed using physical models when all shooting parameters and compatible software are available, or using other mathematical models (rational function model, polynomial model, affine model, direct linear transformation model, grid projection model) when shooting parameters and compatible software are incomplete.

3. Use GPS measurement points or digital elevation models to correct remote sensing images when the displacement of image points caused by terrain height differences is ≥ 0.3mm on the map scale. Otherwise, use the average plane of the entire area to correct the image.

Depending on the modeling method, determine the number and shape of control points. In cases where remote sensing image modeling uses physical models, at least 10 control points must be determined for each model. For polynomial modeling, more than 10 control points evenly distributed across the image must be determined for each model. For physical modeling of adjacent scenes or thickening of remote sensing image blocks, control points must be selected in the overlapping area between the two scenes.

4. Check geometric correction errors:

a) The average positional difference of named objects on corrected remote sensing maps compared to the map should not exceed 0.4mm for clear objects and 0.6mm for unclear objects.

b) Boundary errors between corrected scenes should not exceed 0.7mm in plains and 1.0mm in mountainous areas.

c) Corrected remote sensing images are converted to the VN2000 coordinate system or according to specific requirements.

Article 23. Processing Spectral Data, Merging Images, and Enhancing Image Quality

1. Spectral processing: Remote sensing images must be spectrally processed to display on screen after geometric correction, then saved as digital image data.

2. Merging remote sensing images: To improve image quality for image interpretation and object analysis, full-color and multi-spectral images that have been color-processed should be merged to create high-resolution color images.

3. Enhancing remote sensing image quality: When images do not meet the required sharpness, quality enhancement is necessary to clearly show the elements. Enhancement is performed channel-by-channel using linear path methods, image filtering, cloud-free image combination, interactive adjustment methods, or other methods. The appropriate linear image filtering method should be selected based on the nature of the original image.

Additionally, multi-spectral image quality can be enhanced through color combination solutions implemented according to natural or false color combinations as required.

4. Checking spectral processing and merging remote sensing images

a) Remote sensing images after spectral processing must meet the following requirements:

- Remote sensing images after spectral processing must achieve good image quality with average contrast and no color bias;

- Colors must be uniform across adjacent scenes;

- The grayscale distribution histogram after spectral processing must fully utilize the range of pixel values.

b) Merged remote sensing images must satisfy the following requirements:

- The images used must not be taken more than a certain time apart;

- The positional error between identical objects on different remote sensing images must not exceed 0.5 pixels of the lower resolution image;

- The resolutions of the remote sensing images must not differ by more than four times.

Article 24. Quality Inspection of 3A Remote Sensing Images

After production, 3A remote sensing images must undergo quality inspection as follows:

1. Coordinate system and projection zone inspection according to requirements.

2. Inspect spectral uniformity.

3. Inspect the accuracy of 3A remote sensing images against control points and check points.

4. Inspect the accuracy between full-color and multi-spectral images, with positional errors not exceeding 1 pixel for scenes captured at the same time.

Chapter VII

TECHNICAL PROVISIONS FOR PRODUCING 3B REMOTE SENSING IMAGES

Article 25. Input Data Entry

3A remote sensing images that meet quality requirements after inspection are entered into the remote sensing image processing system as input data for producing 3B remote sensing images.

Article 26. Merging Images and Cutting Map Sheets of Remote Sensing Images

1. Merging remote sensing images: In cases where map sheets of remote sensing images are not entirely within one scene but span multiple scenes, these scenes must be merged.

2. Cutting map sheets of remote sensing images: Remote sensing map sheets are cut according to terrain map fragments or predefined boundaries.

3. Checking merging and cutting:

Remote sensing map sheets after merging and cutting must be inspected:

a) The seam of the merged images must not pass through control points. It must go through objects with the smallest alignment errors, avoiding cutting along linear features. The angle between the seam and linear features must be between 30°-150°;

b) The alignment error when joining images at the boundary of objects on corrected images must not exceed 0.6mm in flat areas and 1.0mm in mountainous regions on the map scale;

c) The tone on both sides of the seam must be relatively uniform. Tone equalization must be performed in a 60-pixel-wide strip around the seam;

d) Remote sensing map sheets must be cut according to terrain map fragments or predefined boundaries.

Article 27. Framework, Coordinate Grid, Explanations, and Content Outside the Frame of Remote Sensing Image Maps

1. The framework and presentation of remote sensing image map frames shall be carried out according to the following general provisions:

a) The remote sensing image map frame shall have geographic coordinates and orthogonal coordinates;

b) The name and designation of the remote sensing image map sheet;

c) The scale of the remote sensing image map;

d) The seam line table;

đ) Information about the remote sensing images (type of image, resolution, time of capture);

e) The name of the unit and the date of creation of the remote sensing image map;

g) The reference system of the remote sensing image map.

2. Checking the framework and presentation of the remote sensing image map frame:

The framework and presentation of the remote sensing image map frame must be checked in accordance with the regulations.

 The template for the frame is specified in Appendix 04 attached to this Circular.

Article 28. Quality Inspection of 3B Remote Sensing Images

After production, 3B remote sensing images must undergo quality inspection.

1. Inspect any stitching marks if present.

2. Check the uniformity of the spectrum with the stitched pieces if present.

Chapter VIII

REGULATIONS ON PACKAGING AND HANDOVER OF PRODUCTS

Article 29. Regulations on Product Acceptance Testing

All remote sensing image products after processing must undergo acceptance testing at the production unit to ensure they meet quantity and quality requirements.

Article 30. Packaging Regulations for Products

1. Remote sensing image products are released in standard formats depending on the type of satellite or in TIFF/GEOTIFF format for common software and other formats as required. Additionally, there must be files containing coordinate information and files containing image information.

2. Digital products are recorded on CD-ROMs or DVDs. CD-ROMs or DVDs recording digital product data must be high-quality discs that can only be written once. Data must not be interrupted during the recording process, and after completion, the content must be rechecked on a computer to confirm the quality of the recorded information. On the cover and label of CD-ROMs or DVDs storing digital products, technical information necessary for storage and data exploitation must be clearly marked, including the designation of the remote sensing image map sheet, scale, date of image capture, name of the unit, and processing time.

3. Printed products on paper must be arranged in sequence and packed in durable material packaging. On the packaging, the product type, scale, quantity of products, implementation time, construction unit, and other necessary notes must be marked.

Article 31. Handover Regulations for Products

1. Remote sensing image products, after undergoing technical inspection, quality evaluation, and acceptance, must be handed over for preservation, utilization, and storage.

2. When handing over remote sensing image products, a handover record must be established. This record serves as the basis for acceptance and settlement of contracts and projects.

Chapter IX

IMPLEMENTING PROVISIONS

Article 32. Effectiveness

This Circular takes effect from May 12, 2015.

Article 33. Implementation Organization

1. Ministries, ministerial-level agencies, agencies under the Government; People's Committees of provinces and centrally governed cities, and related organizations and individuals are responsible for implementing this Circular.

2. The National Remote Sensing Center, Ministry of Natural Resources and Environment shall be responsible for guiding and supervising the implementation of this Circular.

3. Provincial Departments of Natural Resources and Environment are responsible for assisting the People's Committees of provinces and centrally governed cities in implementing this Circular at the local level.

4. During the implementation process, if there are any difficulties, agencies, organizations, and individuals are requested to promptly report to the Ministry of Natural Resources and Environment for research and amendment./.