Attachments_A_B_and_C.pdf

APFO SPECIFICATION FOR DIGITAL CAMERA BASED ACQUISITION ATTACHMENT A

AERIAL PHOTOGRAPHY FIELD OFFICE (APFO)

SPECIFICATION FOR DIGITAL CAMERA BASED ACQUISITION

(Dated February 1, 2012 – Modified for USFS Resource Imagery March 26, 2012)

1.0

INTRODUCTION AND BACKGROUND

The U.S. Federal Government has not established an independent government evaluation and calibration policy for digital cameras since sensor technology is still rather new. Until a policy is developed and implemented, the U.S. Department of Agriculture (USDA), Farm Service Agency (FSA) has proceeded to validate the quality and capabilities of current digital cameras by obtaining relevant information from camera manufacturers, data providers, and other government agencies and organizations. The following specifications and requirements have been developed to ensure that any digital camera proposed for use on USDA contracts meets minimum requirements to provide the highest quality digital imagery products.

2.0

DIGITAL CAMERA SPECIFICATIONS AND REQUIREMENTS

This document covers camera specifications and requirements for any direct-digital imagery acquisition under contract to APFO. Cameras for acquiring precise vertical digital imagery are required to be tested and calibrated. Cameras proposed for use must be of comparable, or better, precision and quality as traditional film-based stereoscopic mapping cameras.

Imagery captured with digital cameras must also be compatible with analytical mensuration procedures used in photogrammetric surveys and in preparing accurate orthophotography.

Documentation and sample imagery will be reviewed and verified by the Government before approval is granted. Only approved digital cameras shall be used on contracts administered by APFO.

3.0

GENERAL REQUIREMENTS

Digital cameras must be tested and calibrated with manufacturer certification documentation.

The camera must be geometrically stable and suitable for use in precise, high-accuracy photogrammetric orthoimagery applications. All delivered imagery shall be acquired and processed in such a way as to eliminate or minimize pixel or band offset or misalignment between bands. The camera shall provide the following:

3.1

Spatial Resolution

The camera shall provide the spatial resolution and field of view necessary to meet the ground sample distance (GSD) requirement as specified in the contract.

3.2 Image Fusion

Pan sharpening will be permitted to achieve the necessary spatial resolution requirements. The multi-spectral bands may be used at a ratio no greater than 1:5 (multi-spectral to panchromatic) to achieve the required spatial resolution.

3.3

Radiometric Resolution and Accuracy

The camera’s sensor shall capture and record a minimum of 12-bits of image information per color channel. If more than one lens and more than one shutter are used in the camera, the difference in radiometric values between two panchromatic or two multi-spectral sensors shall be less than ±5%. For example, a 12-bit image shall not have more than ±205 difference in gray values.

3.4

Spectral Resolution

The camera shall capture, as a minimum, natural color (approximately 440 – 850 nm) and near infrared color (approximately 780 – 850 nm) channel data simultaneously or near simultaneously using a single camera (near simultaneously is defined as less than 500 milliseconds). Additional multi-spectral bands may be collected with a secondary or auxiliary camera and/or system.

3.5

Camera Operation

The digital camera and its mount shall be checked for proper installation prior to each mission. An automatic exposure control device is permitted, but a manual override capability is required for some types of terrain to achieve proper coverage and exposure. The camera mount shall be regularly serviced and maintained and shall be insulated against aircraft vibration.

3.6

Camera Maintenance

The contractor shall perform all maintenance in accordance with the manufacturers recommended and established procedures. The contractor shall maintain a complete history of all maintenance done to the camera system and have it available for Government inspection. The contractor shall provide certification that the system has been maintained, preventive maintenance and calibration performed, to the manufacturers requirements.

3.7

System Malfunctions

The contracting officer shall be notified of all camera malfunctions within 72 hours with a written report of the malfunction. A malfunction is defined as a failure in any element or process of the camera that causes an interruption of the normal operations of the camera system (camera system is defined as the camera and any key components, such as camera mount, airborne global positioning system, and on-board data storage).

All malfunctions or failures of global positioning systems or inertial measurement unit systems shall also be reported directly to the contracting officer.

4.0

DIGITAL CAMERA APPROVAL REQUIREMENTS

All digital cameras must be approved by the Contracting Officer before acquiring imagery under any APFO contract. When requesting approval, the Contractor shall submit, or have on file with APFO, a report of calibration (see Paragraph 4.1), sample digital imagery (see Paragraph 4.2), and camera documentation (see Paragraph 4.3).

4.1

Calibration Reports

Calibration reports for each digital camera proposed for use shall be submitted to the contracting officer with the contractor’s proposal and prior to project imagery acquisition if the digital camera is removed and remounted. The contractor shall follow manufacturer’s specifications for appropriate calibration and recalibration. The calibration reports shall address the geometric performance of the camera, and at a minimum, include:

(a) Date of report

(b) The name of the person or company performing the calibration

(c) The methodology and procedures used for calibration

(d) Final calibration parameters, such as calibrated focal length, lens distortion values, radiometric calibration parameters, and principal point location.

NOTE: The government recognizes that individual calibration reports, procedures, and parameters may be unique to a certain manufacturer since equipment and systems vary from manufacturer to manufacturer.

4.2

Sample Imagery Requirements

The Contractor shall acquire and submit with their proposal, sample images from the digital camera proposed for use. The sample imagery shall provide the following minimum characteristics:

(a) Display the same GSD resolution required in the solicitation.

(b) Represent the type of terrain (agriculture, cropland, forest, etc.) that is similar to the proposed project area.

(c) Re-sampled and submitted as an 8-bits per band image (unless the solicitation requires only 16-bit per band image delivery, in which case the sample imagery shall be submitted as a 16-bits per band image).

(c) If ortho-rectification is required under the proposed solicitation, the sample image shall be ortho-rectified with the projection specified in the solicitation (for example, North American Datum 1983 (NAD83) and UTM Zone 12).

(d) Sample shall be produced and submitted in the footprint and file format specified in the solicitation (for example, DOQQ formatted, GeoTIFF image).

(e) The sample imagery shall fit on one standard CD or DVD. Delivered media will become part of the official Government contract file and will not be returned.

4.3

Camera Documentation Requirements

The Contractor shall provide with their proposal detailed documentation of the digital camera proposed for use. Documentation may include brochures, technical specifications, marketing material, manufacturer’s user manuals, or other descriptive literature. The documentation shall contain at a minimum the following information:

(a) General overview information

(b) Product configuration description

(c) Camera component description

(d) Technical specifications

(e) Computer management and storage systems

(f) Image acquisition and processing workflow

DIGITAL IMAGERY QUALITY SPECIFICATION ATTACHMENT B

AERIAL PHOTOGRAPHY FIELD OFFICE (APFO)

USDA DIGITAL IMAGERY QUALITY SPECIFICATION

(Dated February 1, 2012 – Modified for USFS Resource Imagery March 26, 2012)

1.0

SCOPE

This document establishes the image quality criteria to be used in the production of digital imagery products for all contracts issued by the United States Department of Agriculture’s (USDA) Aerial Photography Field Office.

2.0

APPLICABLE DOCUMENTS

In the event of conflict between the contents of this specification and the documents referenced herein, the contents of this specification shall take precedence.

2.1 National Agriculture Imagery Program (NAIP) Suggested Best Practices – Final

Report, dated Feb 1, 2007 (ITT Space Systems Division)

3.0

GENERAL REQUIREMENTS

USDA uses imagery for various programs including, but not limited to forest management, agriculture land use analysis, natural resource inventory, and extraction of data by means of photogrammetric interpretation. The complex nature and need for consistent imagery requires adherence to exact format and content of this specification.

3.1 Image blemishes, scratches and artifacts

. Imagery shall be free of blemishes, scratches, and artifacts that obscure ground feature detail. The following table defines the maximum acceptable limits for blemishes, scratches, and artifacts. Clusters of blemishes, scratches, and artifacts that do not individually meet these criteria may be considered unacceptable.

ACCEPTABLE, IMAGE BLEMISHES,

SCRATCHED, AND ARTIFACTS

1 pixel wide 100 pixels in length 2 pixels wide 60 pixels in length 3 pixels wide 20 pixels in length

4 – 12 pixels wide 12 pixels in length

3.2 Band-to-Band Registration Accuracy

. Misregistration between any color bands shall not exceed 1 pixel.

3.3 Original Image Resolution

. The original image, original scan, or original capture used to create the imagery shall not be resampled from the original image resolution greater or less than the following numbers in order to meet the Ground Sample Distance (GSD) specified in the contract:

GROUND SAMPLE ORIGINAL IMAGE RESOLUTION

DISTANCE (GSD) MAXIUMUM (meters) MINIMUM (meters)

0.3-meter 0.15 0.32

0.5-meter 0.25 0.53

1-meter 0.50 1.05

4.0

UNCORRECTED IMAGERY

Uncorrected imagery is defined as imagery that has been minimally processed before exporting to a non-camera specific file format, such as a TIFF. Uncorrected imagery is the closest “match” to a traditional film negative that the direct-digital camera can provide without having the end-user employ special and/or non-standard software.

4.1 Non-image data

. Imagery shall only use a pixel digital number (DN) of zero (0) for non-data values.

4.2 Image Quality

. The Contractor shall not make any radiometric enhancements, such as gamma correction, histogram stretching, dodging, or other Look Up Table (LUT) adjustments, to the acquired imagery. The imagery shall not contain any borders, artifacts, or other non-image items.

5.0

COLOR CORRECTED IMAGERY

Imagery required to be color-corrected shall be adjusted so that the image matches the ground at the time of exposure. Adjustments shall include, but not limited to, any dodging, gamma correction, histogram stretching, brightness adjustments, and/or color balancing. The files shall not contain any borders, artifacts, or other non-image items.

5.1 Non-image data

. Imagery shall only use a pixel digital number (DN) of zero (0) for non-data values.

5.2

Natural Color Image Quality.

(a) Clipping. Imagery shall have a tonal range that prevents the clipping of highlight or shadow detail from the image. When calculated against the luminosity histogram, the cumulative pixel count between the first and last five histogram bin values (5 and 250 respectively for 8-bit depth) shall not be less than 98.0%, with a preferred value greater than 99%.

(b) Contrast

. When calculated against the luminosity histogram, the difference between the histogram digital number (DN) value that contains 99.0% of the cumulative pixel count and the DN value that contains 1.0% shall be greater than ±59% of the bit depth, ±4% (aim point of 150, ±10 for 8-bit depth). If the cumulative pixel count percentage falls between two histogram bin values, the closest value shall be used. For example, if an 8-bit image has a luminosity DN value 222 contains 99% of the cumulative pixel count and DN value 44 contains 1% count, therefore the difference is 178.

BIT DEPTH DN DIFFERENCE

TARGET MINIMUM MAXIUMUM

8-bit 150 140 160 16-bit 38,550 35,930 41,170

(c) Brightness

. Imagery shall have a mean pixel count within ±7.5% of the middle DN value allowed for the bit depth. For example, an 8-bit depth image must have the histogram mean value between 108 and 147.

BIT DEPTH MEAN DN

MINIMUM MAXIUMUM

8-bit 108 147 16-bit 27,853 37,683

(d) Color Balance

. Imagery should have a neutral tonal range without the dominance of any individual color. The difference between the minimum and maximum DN value in a RBG triplet of any nearly neutral objects within the image shall be less than 5.

5.2 Color infrared Imagery

. All color infrared imagery shall have proper contrast to allow highlight and shadow detail.

5.3 Multispectral Imagery

. Multispectral Imagery shall be radiometrically processed such than the natural color bands (RGB) meet the quality requirements in paragraph 5.1.

6.0

ORTHORECTIFIED IMAGERY

All orthorectified imagery shall be color-corrected in accordance with paragraph 5.0.

6.1 Geographic Extent. Imagery shall cover the entire image area, including the required minimum buffer on all four sides. Extents shall be computed by projecting the geographic corners and side midpoints to the appropriate projection, then adding the buffer on each side of the resulting minimum bounding rectangle.

6.2 Specular reflections

. Specular reflections in the imagery should be minimized, especially in agriculture areas, by patching the area using “chips” from different imagery but shall be from the same type of camera and must be from the same acquisition season. Any chips used in the imagery shall not have more than ±3 pixels offset or the specified horizontal accuracy, whichever is the lesser distance, between the chip and principal image. Any chips used shall be radiometrically balanced in accordance with paragraph 6.3(a).

6.3 Image Mosaicking

. Imagery may be created using multiple image segments from the same acquisition collection to produce the final product.

(a) Radiometric Balance

. When a mosaic is made from two or more image segments, the brightness and color values between the image segments will be adjusted to match that of neighboring image segments. The join lines between the overlapping image segments will be chosen to minimize tonal variations.

Localized adjustment of the brightness and color values will be done to reduce radiometric differences between join areas.

6.4 Spatial

. All orthoimagery shall meet the horizontal accuracy requirement of the specified project.

7.0

DEFINITIONS

Chip – Each separate piece of a mosaicked image that contributes to the final image.

Clipping – The presence of pixels exhibiting the minimum or maximum digital number in an image’s dynamic range.

Digital Number – The value (0-255 for an 8-bit image) that depicts the pixel radiance for that color band.

Dodging – Manipulating the intensity of part of a photograph by selectively shading or masking.

Resample – Interpolation of pixel values based upon neighboring pixel values.

Uncorrected Imagery – Imagery that has been minimally processed, including no radiometric enhancements, such as stretching, dodging, or other Look Up Table (LUT) adjustments, to the acquired imagery.

USDA-Farm Service Agency Aerial Photography Field Office

2222 West 2300 South Salt Lake City, UT 84119

USDA Digital File Format Specification

Version 1.0 01/24/2013

Page i

Contributing Organizations

USDA

Document contributor contacts Author Agency Contact Information

APFO USDA-FSA john.mootz@slc.usda.gov

Revision History

Date Version Editor Outline Number modified

Description of Change

1/24/13 1.0 Vanderbilt N/A N/A

Table of Contents

1.0 Preface

2.0 Introduction

3.0 Scope

4.0 Conformance

5.0 Normative Documents

5.1 Industry Specifications

5.2 National Specifications

5.3 Informative References

6.0 Glossary

6.1 Terms and Definitions

6.2 Acronyms and Abbreviations

6.3 File Extensions

7.0 Applicability and Use of this document

8.0 TIFF and GeoTIFF Requirements

8.1 Data Capture

8.2 TIFF and GeoTIFF Requirements

9.0 LizardTech’s MrSID Requirements

10.0 JPEG 2000 Requirements

11.0 ERDAS IMG® Requirements

12.0 ERDAS ER Mapper ECW® Requirements

Appendix A

1.0 PREFACE

For many years the Aerial Photography Field Office (APFO) has used the “USDA Digital Orthoimagery

Quarter Quadrangle (DOQQ) Description and Specification” when contracting for Farm Service Agency’s

(FSA) National Agriculture Imagery Program (NAIP) imagery. However, APFO is the lead procurement office for all United States Department of Agriculture’s (USDA) imagery projects greater than 100 square miles. These programs require a variety of delivery and file formats to meet the needs and diversity of various USDA’s users. The previous specification was written specifically for one program, and thus was too rigid to be easily implemented for other acquisitions, resulting in continuous modifications to the specification and inefficiencies in implementation, interoperability, and data exchange. Thus, this specification has been updated to alleviate these issues. The intent is to apply this specification, as written and without modification, to all imagery contracts.

An important area of consideration for the use of imagery (raster) data is the encoding format. Various department agencies that use the APFO for procurement of imagery have valid requirements to make selected imagery data holdings available in Geographic Tagged Image File Format (GeoTIFF) in addition to other formats, including but not limited to JPEG 2000, LizardTech MrSID, and ERDAS Imagine files.

While GeoTIFF (and other formats) are widely used, they have many format options which often result in non-interoperability among disparately developed implementations. This USDA Digital File Format

Specification was developed to help meet objectives for deployment of capabilities with automation and interoperability in mind. The primary rationale is to enable more timely support for delivering imagery in a format suitable for direct ingest by Geographic Information System (GIS) and/or Remote Sensing

(RS) application software tools readily available to those working to support department activities.

This document specifies the requirements that shall be used for the exchange of georeferenced or ortho rectified imagery when opting to use the varied formats identified in the Appendices. The aim of this specification is to enable the interchange of rectified constraints within the design objectives of promoting interoperability for the exchange of raster information.

2.0 INTRODUCTION

This specification outlines the requirements and encoding rules that shall be used for the exchange of imagery data when opting to use the formats outlined herein. It constitutes a description of the bounds and constraints for the use of each file format within the design objectives of promoting interoperability and data exchange. It conforms with all normative documents identified in Section 5.0.

While GeoTIFF is not the only format detailed in this specification, it is a core format for encoding spatial raster data, and thus is a core component of this specification. GeoTIFF is a public domain specification which allows georeferencing information to be embedded within a TIFF file. Potential embedded information includes projections, coordinate systems, datums, and other information needed to establish the spatial reference for the imagery or gridded data contained in the file. The GeoTIFF format augments the TIFF format, so TIFF-enabled software incapable of reading and interpreting the specialized georeferencing metadata should still be able to open a GeoTIFF file sufficiently to at least view the image data.

The main body of this specification addresses the scope of this work, conformance with other recognized industry specifications and national specifications, terms and definitions, an acronym and file extension reference, and the applicability and use of this document. The document then addresses various file format requirements and references the Appendices for the file format specifications.

3.0 SCOPE

This document establishes the file format criteria to be used in the production of digital imagery for all contracts issued by APFO. It may also be applied to all internal production activities at APFO that result in products that require archiving and public distribution.

4.0 CONFORMANCE

Any GeoTIFF data claiming conformance with this specification is claiming conformance with applicable normative documents in Section 5, and shall pass conformance requirements testing as part of the quality assurance process upon delivery of data to the USDA.

Other file formats detailed in this specification may or may not conform with normative documentation and reference information in Section 5. These file formats shall still pass testing as part of the quality assurance process upon delivery of data to USDA, to ensure requirements of this specification are met.

5.0 NORMATIVE DOCUMENTS

The following referenced documents are indispensable for the application of this document. For dated references, only the edition cited applies.

5.1 Industry Specifications

The Tagged Image File Format (TIFF) – Tagged Image File Format, Revision 6.0, Adobe

Systems Inc., June 1992. TIFF is a copyrighted standard of Adobe Systems, Inc.

(http://partners.adobe.com/public/developer/en/tiff/TIFF6.pdf)

GeoTIFF Format Specification, Revision 1.0, 10 November 1995; version 1.8.2 The GeoTIFF

Format Specification is a public domain extension of TIFF that provides a robust and flexible method of storing georeferencing information is a TIFF file.

(http://www.remotesensing.org/GeoTIFF/spec/GeoTIFFhome.html ) http://partners.adobe.com/public/developer/en/tiff/TIFF6.pdf http://www.remotesensing.org/geotiff/spec/geotiffhome.html

Supporting documents describing the scope of this document and call out requirements, and recommendations.

5.2 National Specifications

TBD

5.3 Informative References

NGA Standardization Document - Implementation Profile for Tagged Image File Format (TIFF) and

Geographic Tagged Image File Format (GeoTIFF) - NGA.IP.0001_1.0, 2008-11-18

6.0 GLOSSARY

6.1 Terms and Definitions

Affine Transformation – A geometric transformation that scales, rotates, skews, and/or translates images or coordinates between any two Euclidean spaces. It is commonly used in GIS to transform maps between coordinate systems. In an affine transformation, parallel lines remain parallel, the midpoint of a line segment remains a midpoint, and all points on a straight line remain on a straight line. Source: Esri Online Dictionary

Band – A range of wavelengths of electromagnetic radiation. Also, image data gathered at this wavelength range.

Field – Refers only to the entire field, including the value, of the geokey (as defined in the TIFF

Specification).

Georeference – Aligning geographic data to a known coordinate system so it can be viewed, queried, and analyzed with other geographic data. (Source: Esri® Online Dictionary)

GeoTIFF – Public domain metadata standard which allows georeferencing information to be embedded within a .TIFF file.

Image File Directory – Contains information about the image. There must be at least 1 IFD in a TIFF file and each IFD must have at least one entry.

Metadata – Description of the content, quality, condition, and other characteristics of the data.

Orthorectification - A process of removing sources of distortion from an image and correcting aerial photography to the point where measurements of a feature on the photograph approximate ground measurements of the same feature. (Source: Satellite Imaging Corporation)

Private tags – TIFF tags numbered 32,768 or higher. Private tags are not defined in the TIFF

Specification.

Public tags – TIFF tags that are defined by the TIFF Specification.

Rectified Grid – A grid that has an affine transformation between the grid coordinates and the coordinates of an external coordinate reference system.

Tag – Refers only to the identifying number portion of the geokey (as defined in the TIFF

Specification).

6.2 Acronyms and Abbreviations

GeoPDF – A .pdf file created by TerraGo Technologies that contains geospatial information such as layers and object data.

ERDAS – Earth Resources Data Analysis System

JPEG and JPEG 2000 – Joint Photographic Experts Group. A commonly used method of compression for digital imagery. JPEG 2000 is an updated wavelet compression of JPEG with more flexibility in the codestream.

MrSID – Multi-resolution seamless image database. Proprietary format developed by LizardTech for encoding raster graphics.

TIFF – Tagged Image File Format.

6.3 File Extensions

.ecw – Stands for enhanced compression wavelet. An ERDAS ER Mapper compressed file format specifically designed for geospatial imagery.

.ige – Stands for large raster spill file. One of two ERDAS IMAGINE files created when an image requiring more than 4GB of disk space is created. It contains the actual image data in a separate non-HFA file format (normally with the extension .ige).

.img – Stands for “image file.” An ERDAS IMAGINE file used to store raster data, including file information, ground control points, sensor information, layer information, attribute data, statistics, map information, projection information, pyramid layers, data file values, compression, and block size. This file uses Hierarchal File Format (HFA).

.jp2 – Stands for jpeg 2000. Compressed image created using JPEG 2000 Core Coding; incorporates a wavelet compression algorithm instead of Digital Cosine Transform (DCT) compression used by standard JPEG images; can be stored using lossy and lossless compression.

.pdf – Stands for portable document format. For the purpose of this document, .pdf refers to a geopdf.

.ovr – Stands for overlay file. Similar to the .rrd file, .ovr is a file format for storing pyramid layers for a raster dataset.

.rrd – Stands for “reduced resolution dataset.” A file containing pyramids created for a raster dataset.

.sid – Stands for seamless image database. A lossless compressed image file developed by

LizardTech

.tif – Stands for “tagged image file.” High-quality graphics format often used for storing images with many colors, such as digital photos; short for "TIFF;" includes support for layers and multiple pages. For the purpose of this document, .tif refers to a GeoTIFF.

7.0 APPLICABILITY AND USE OF THIS DOCUMENT

The primary purpose of this document is to establish guidelines and specifications for imagery developed by or for APFO. The intent is to allow for great flexibility of data structure, spatial density, quality, and format, while supporting sharing and reuse of the data.

8.0 TIFF AND GEOTIFF REQUIREMENTS

This section of this File Format Specification defines the minimum set of format options, features, and data elements necessary to promote interoperability for the exchange of Tagged Image File

Format (TIFF) and GeoTIFF-formatted files. This section establishes a common frame of reference to address adequate georeferencing parameters and data value attribution needed by geospatial applications to utilize GeoTIFF-formatted images. Only uncompressed TIFF or GeoTIFF imagery that has been georeferenced or orthorectified is addressed by this section of this specification.

Requirements for compression (see Sub-Section 8.2.10) and georeferenceable imagery (see Sub-

Section 8.2.2) are not supported in this specification.

TIFF is an image file format used for storing and interchanging raster images. TIFF is a portable format, not specific to or favoring any particular operating systems, file systems, compilers, or processors. It is also extensible and designed to evolve as new needs arise.

GeoTIFF is a set of TIFF tags (extensions to the Baseline TIFF Format) that describe spatial information associated with TIFF imagery that originate from any digital sensor, scanned aerial photography, scanned maps, or as a result of geographic analysis or similar processes. GeoTIFF uses a small set of reserved TIFF tags to store a broad range of geo-referencing information.

This specification addresses the following fundamental topics for application and use of TIFF and

GeoTIFF files:

Required TIFF and GeoTIFF tags for standardized use of TIFF and GeoTIFF-formatted files.

Image data or pixel structures relating to: strips, tiling, compression, precision, number of bands, etc.

Preferred coordinate reference systems, datums.

This specifications main characteristic allows:

All types of imagery conformant to TIFF Baseline: bi-level, grayscale, palette color image, 3-bands RGB (full color), and multi-spectral.

Optional TIFF tiling (TIFF extension specified in section 15 of the TIFF standard) for high volume data. This option cannot be used in conjunction with TIFF striping. The use of this option may cause some interoperability problems, and must be identified as such at a different conformance level.

8.1 Data Capture

TIFF provides 2 tags for specifying the scanner or instrument manufacturer and model: Make and

Model. These fields may be populated according to product specification requirements.

However the production process is usually far more complex than the simple acquisition of an image by a scanner or a camera, and the full process needs to be documented by additional metadata (i.e. process description).

Use of Make and Model tags is consequently optional: in case they are populated, the information should be consistent with additional metadata (i.e. process description).

8.2 TIFF and GeoTIFF Requirements

In general, imagery data delivered in TIFF or GeoTIFF format will conform to (not deviate from) the

TIFF and GeoTIFF formats as defined in the referenced TIFF and GeoTIFF specifications. The following clauses constrain the implementation of TIFF and GeoTIFF for the purposes of this specification.

8.2.1 General File Structure and Data Value Types

The TIFF structure includes an 8-byte image file header that points to the first Image File Directory

(IFD). This specification calls for a single IFD in each TIFF file and this IFD must have at least one entry. The IFD contains information about the image, as well as pointers to the actual data. All used fields are listed, and information that does not fit in the IFD is listed in other parts of the file.

The IFD begins with a 2-byte count of the number of directory entries (i.e., the number of fields), followed by a sequence of 12-byte field entries, followed by a 4-byte offset of the next IFD (or 0 in the case there is no other IFD). This 4-byte (32 bit) structure of the offset allows TIFF and GeoTIFF file sizes up to 4GB (2**32 bytes). This specification only supports file sizes up to 2GB.

A GeoTIFF file is a TIFF 6.0 file, and inherits the file structure as described in the corresponding portion of the TIFF Specification. All of the GeoTIFF information is encoded in six TIFF tags, which are designed to store a broad range of georeferencing information, catering to geographic as well as projected coordinate system needs. These GeoTIFF keys will contain no private Image File

Directories (IFD's), binary structures, or other private information invisible to baseline TIFF 6.0 readers.

The GeoTIFF 1.0 standard uses a MetaTag (GeoKey) approach to encode dozens of data elements into just six TIFF 6.0 tags. GeoKeys are structurally similar to TIFF 6.0 tags, but at one lower level of abstraction. GeoKeys are used within the tags to store the projection parameters and coordinate system information. All keys are referenced from one tag, the

GeoKeyDirectoryTag. See Table 2.1, Appendix A for details. Not all the keys will be used when formatting GeoTIFF in conformance with this specification. This specification requires that only certain essential GeoTIFF keys be populated. Other keys may require default values.

The GeoTIFF specification requires interpret (reader) implementations to support all documented

TIFF 6.0 tag data-types, and in particular requires the Institute of Electrical & Electronic Engineers

(IEEE) double-precision floating point ‘DOUBLE’ type tag. The documented data types for use with

TIFF tags are:

BYTE = 8-bit unsigned integer

ASCII = 8-bit byte that contains a 7-bit American Standard Code for Information Interchange

(ASCII) code. The last byte of an ASCII sequence (string) must be null (binary zero)

SHORT = 16-bit (2-byte) unsigned integer

LONG = 32-bit (4-byte) unsigned integer

FLOAT = Single precision (4-byte) IEEE format

DOUBLE = Double precision (8-byte) IEEE format

RATIONAL = Two LONGs: the first represents the numerator of a fraction; the second, the denominator

SBYTE = 8-bit signed (twos complement) integer

UNDEFINED = 8-bit byte containing anything, depending on the definition of the field.

SSHORT = 16-bit (2-byte) signed (twos complement) integer

SLONG = 32-bit (4-byte) signed (twos complement) integer

SRATIONAL = Two SLONGs: the first represents the numerator of a fraction; the second, the denominator.

Note: Appendix A identifies which data type applies to each tag selected for use by this specification.

TIFF implicitly types all range values (data sample values) as unsigned integer values. The representation of imagery, however, requires the ability to store the range (data) values in additional representations such as signed integer and floating point. Section 19 of the TIFF standard presents a scheme for describing a variety of data sample formats. The BitsPerSample field in the TIFF Image File Directory defines the number of bits per component.

8.2.2 Rectification and Ortho-rectification

A rectified grid has an affine transformation between the grid coordinates and the coordinates of an external coordinate reference system. A rectified grid is defined by an origin in an external coordinate reference system and a set of offset vectors that specify the direction and distance between grid lines within that external coordinate reference system. If the coordinate reference system is related to the earth by a datum, the grid is a georeferenced grid. An orthorectified grid is a georeferenced grid that uses elevation data and where constant scale is maintained throughout the grid.

A referenceable grid is one that can be referenced by some other specified coordinate transform

(for example, by a physical sensor geometry model or by a functional fit model of rational polynomials).

This section of this specification is concerned only with georeferenced grids and orthorectified grids. It does not address referenceable grids; for example, those associated with oblique imagery or rubber sheeting.

The GeoTIFF Format Specification describes 5 coordinate transformations ‘cases’ that the format is able to address:

Case 1: The model-location of a raster point (x,y) is known, but not the scale or orientations.

Case 2: The location of three non-collinear raster points are known exactly, but the linearity of the transformation is not known, and the scale is not to be defined.

Case 3: the position and scale of the data is known exactly, and no rotation or sheering of the image is needed to fit into the model space.

Case 4: (intended for equidistant- sampled data) The raster data requires rotation and/or lateral sheering to fit into the defined model space. To accomplish this, additional information is needed in the form of a transformation matrix. The ModelTransformationTag exists for the purpose of allowing this information to be provided.

Case 5: The raster data cannot fit into the model space with a simple affine transformation.

Multiple tie points can be stored in GeoTIFF to allow rubber sheeting of the image in this case.

This specification requires that the scale be identified in the GeoTIFF tags (see Table 2.1 Appendix

A), and therefore cases 1 and 2 are not supported. This specification requires only georectified grids and orthorectified grids, therefore case 5 is not supported.

Only cases 3 and 4 are supported by this section of this specification.

8.2.3 Coordinate Reference Systems and Datums

The GeoTIFF Configuration GeoKeys establish the general configuration of the file’s coordinate system. Each of these GeoKeys is listed below with their general description followed by limitations and constraints established by this specification:

GTModelTypeGeoKey – Tag 1024. The GTModelTypeGeoKey defines the general type of model coordinate system used – geographic or projected.

GTRasterTypeGeoKey– Tag 1025. The GTRasterTypeGeoKey establishes if the raster pixel value (imagery or gridded data range value) is located at a point value or if the value fills the square grid cell.

GTCitationGeoKey – Tag 1026. The GTCitationGeoKey is used to give an ASCII reference to published documentation on the overall configuration of the GeoTIFF file. This key generally shows the projection name or geographic coordinate system name and the units.

Horizontal datum – GeoTIFF has many datums to choose from in the Geodetic Datum numerical codes. This specification recommends the use of North American Datum (NAD83) as the horizontal datum, but allows for World Geodetic System 1984 (WGS84).

Coordinate systems – GeoTIFF allows many geographic latitude-longitude systems and many projected coordinate systems. There are also tags for parameters of coordinate systems or projections not available in the codes. This section of this specification limits expression of coordinate references to NAD83 latitude and longitude (decimal degrees), NAD83 UTM Grid

System Northing and Easting (meters), and geographic or projected WGS84 systems.

8.2.4 Units of Measure

This section of this specification allows only the implied ‘angular degree’ unit for the geographic coordinate system, and the default ‘meters’ unit for the UTM projected coordinate system. User-defined geographic or projected coordinate systems are not allowed by this specification. In order to prevent the use of other horizontal units of measure, the use of the GeoKeys related to horizontal units of measure are prohibited (see Tables 2.x in Appendix A).

Default units are:

Decimal degrees for longitude and latitude (geographic coordinate system)

Meters for UTM Grid System Easting and Northing (projected / cartographic coordinate system)

8.2.5 Date and Time

There is a TIFF field called DateTime for storing the date and time of file creation. The format for the field in ASCII type is “YYYY:MM:DD HH:MM:SS” with 24 hour time used for the hours and one space character between the date and time, and one terminating NUL character. The length of the string, including the terminating NUL, is 20 bytes. All dates and times shall be expressed in

Coordinated Universal Time (UTC).

Use of this tag is recommended in order to support discovery of the data, wherever possible. This information should then be consistent with additional metadata (i.e. required process description).

Absence of this tag indicates this information was not available. The Date/Time stamp that will be represented in the TIFF DateTime field shall be the date/time when the image file was created.

8.2.6 Collection and Maintenance Constraints

There are several TIFF tags that can carry and address a variety of collection information. These tags should not be populated for the purpose of this section of this specification. The associated additional metadata (i.e. process description) can optionally be used to carry this type of information when needed.

8.2.7 Tiling

For low-resolution to medium-resolution images, the standard TIFF method of breaking the image into strips is adequate. However high-resolution images (grids greater than 8192 x 8192) can be accessed more efficiently if the image is broken into roughly square tiles instead of horizontally-wide but vertically-narrow strips.

TIFF extensions offer an internal TIFF tiling mechanism which should be used on large grids / images, based on the most common tiling scheme which is a rectangular grid, by specifying additional fields for rectangular tiles (for example width and length of a tile). Tile dimensions must be a multiple of 16 (TIFF specifies TileWidth and TileLength be a multiple of 16 for performance in some graphics environments and compression schemes such as JPEG). This internal TIFF tiling extension may not always be supported by commercial or public domain software, especially older

TIFF readers.

TIFF internal tiling must NOT be used in conjunction with stripping. When using internal tiles, the grid data may need to be padded to tile boundaries when the grid size is not an integer multiple of the selected tile size.

Generally, for small grids, the data should be organized as a single TIFF file with no tiling, in order to maximize interoperability.

For large grids (greater than 8192 x 8192), TIFF tiling becomes a helpful option, the recommended tile size is 1024 x 1024. Another option is external tiling, when each tile is typically stored within separate files; this option is outside the scope of this section of this specification and is therefore not supported.

8.2.8 Number of Bands

The number of bands within a GeoTIFF grid is constrained to be 1-band, 3-bands, and 4 or more bands (multi-spectral). For the case of 3 or more bands, the band interleave shall be the TIFF

‘chunky’ format, band interleaved by pixel (BIP). In Chunky format the component values for each pixel are stored contiguously. For example, for RGB data, the data is stored as RGBRGBRGB. The

BIP data organization can handle any number of bands, and thus accommodates black and white, grayscale, true color, and multi-spectral image data.

Additional information is needed to interpret the image data, such as SamplesPerPixel, BitsPerSample, PhotometricInterpretation, and ExtraSamples:

For 1-band 8-bit Pan Sharpened imagery, the following TIFF fields are documented as follows:

SamplesPerPixel = 1, BitsPerSample = 8, PhotometricInterpretation = 1 (BlackIsZero), Do not use

ExtraSamples.

For 3-band 8-bit RGB imagery, the following TIFF fields are documented as follows: SamplesPerPixel

= 3, BitsPerSample = 8, PhotometricInterpretation = 2 (RGB), Do not use ExtraSamples.

For 4-band 8-bit RGB imagery, the following TIFF fields are documented as follows: SamplesPerPixel

= 4, BitsPerSample = 8, PhotometricInterpretation = 2 (RGB), ExtraSamples = 0 (Unspecified data).

For 8-band 16-bit multi-spectral imagery, the following TIFF fields are documented as follows:

SamplesPerPixel = 8, BitsPerSample = 16, PhotometricInterpretation = 2 (RGB), ExtraSamples = 0

(Unspecified data).

8.2.9 Range Value Data Types and Precision

For imagery, the range (data) values are constrained to be unsigned integer data, 8 or 16-bits-per-band.

8.2.10 Compression

No TIFF (internal) compression shall be used.

8.2.11 Image File Implementation

This specification addresses the following cases (except explicit contrary notice): base (single) image (one single IFD). The IFD always addresses the image data.

8.2.12 Security Classification

There are no dedicated fields for storing security classification information in TIFF files. Additional metadata should be used to associate security markers and dissemination controls for content of

GeoTIFF files. However, inclusion of security constraint information in the ImageDescription tag is an option proposed by this section of this specification in order to support security marking of the data. This provides a consistent place for security metadata within the TIFF structure.

It is recommended to include security constraint information in the ImageDescription field, in addition to the information provided in the additional metadata (i.e. process description) so that this information is displayed even if this metadata is ignored by an application or become separated. This security constraint information should then be consistent with dedicated additional metadata (i.e. process description).

8.2.13 Data Quality

There are neither fields nor any mechanism for storing data quality information (positional accuracy, currency, quality information etc.) in the GeoTIFF format and in this specification. This information needs to be documented by additional metadata (i.e. process description).

9.0 LIZARDTECH’S MRSID REQUIREMENTS

MrSID imagery shall be compressed and saved in Generation Three (MG3) format. When encoding the image, the following settings shall be applied:

compression block size of 64 both the transparency and background values set to an RGB value of 0,0,0 (black) use the “maximum zoom level” applicable to the input image, for example: -checking the “Use

Maximum Zoom Levels for Image” button in the encoding options menu.

Compression Ratio. Compression ratio shall be determined by the project or contract specifications. All compression shall be at the same ratio and settings ("region of interest" compressed at a different ratio will not be accepted).

Header Information. The image header shall contain correct Esri® compatible projection information for the mosaic.

Required Files. All standard MrSID® MG3 files generated by the LizardTech software (i.e., .sid, .sdw, and

.txt) shall be included.

Configuration File. Provide the text file created when generating the image.

Auxiliary File. Provide an “.aux” file containing Esri® projection information for each image. The auxiliary file shall contain the proper projection information for the image and shall match the information in the image header.

10.0 JPEG 2000 REQUIREMENTS

JPEG 2000 imagery shall be compressed and saved in the JPEG 2000 format with an unsigned, 8-bit depth. Compression ratio shall be determined by the project or contract specifications. When encoding the image, the following settings shall be applied:

Tiling: None

Code blocks: 64

Precincts: 256 x 256

Strip height: 12

Progression order: rpcl

Quality layers: 8

Packet length markers: Yes

Filter: 9-7

Tile length markers: No

Transparency: Yes

Background: Transparent, Black, White (stated in order of preference)

All compression shall be at the same ratio and settings ("region of interest" compressed at a different ratio will not be accepted).

11.0 ERDAS IMG® REQUIREMENTS

IMG imagery shall use the standard outputs as defined in the hierarchical file structure (HFA). The .img file shall be in a tiled format with a block size of 64x64 pixels. The following statistics shall be calculated:

Minimum and maximum data file values

Mean of the data file values

Median of the data file values

Mode of the data file values

Standard deviation of the data file values

If the file has been georeferenced, the following map information shall be stored in the raster layer:

Upper left x, y coordinates

Pixel size

Map unit used for measurement (e.g., meters, inches, feet)

An .ige file shall be created for images larger than 4GB. The .ige file shall be in a non-HFA format containing only the actual image data.

12.0 ERDAS ER MAPPER ECW® REQUIREMENTS

When creating ECW files, the following input settings shall be used:

Output file type is .ecw

Compression type is set to correct spectral resolution (color RGB, grayscale, multiband)

Generate NULL opacity mask channel is checked

Compression ratio shall be determined by the project or contract specifications.

Output resolution is set by the compressor

ERDAS ER Mapper® must be used when compressing imagery to the .ecw format. This ensures that all image keys and tags are correctly populated in the output image header.

Appendix A: TIFF/GeoTIFF Image Format Constraints USDA Digital File Format Specification

APPENDIX A

TIFF/GeoTIFF Image Format Constraints

TIFF Format The file structure for the Image File Header and Image File Directory can be found in the TIFF product specification in Part 1, Section 2: TIFF Structure. There are four Baseline TIFF image types; Bilevel Images, Grayscale Images, Palette-color Images, and RGB Full Color Images. Each type has specific field requirements which can be found in

Sections 3 through 6 of the TIFF document. For the purpose of representing imagery, the grayscale and full-color image baselines are appropriate. The use of any private tags, other than those included in this appendix is prohibited for use.

TIFF Specification Document: Revision 6.0 dated June 3, 1992

Table 1.1 describes the “Baseline Fields” defined in Section 8 of the TIFF specification.

Table 1.2 describes the “CCITT Bilevel Encodings” defined in Section 11 of the TIFF specification.

Table 1.3 describes the “Document Storage and Retrieval” defined in Section 12 of the TIFF specification.

Table 1.4 describes the “Differencing Predictor” defined in Section 14 of the TIFF specification.

Table 1.5 describes the “Tiled Images” defined in Section 15 of the TIFF specification.

Table 1.6 describes the “CYMK Images” defined in Section 16 of the TIFF specification.

Table 1.7 describes the “HalftoneHints” defined in Section 17 of the TIFF specification.

Table 1.8 describes the “Associated Alpha Handling” defined in Section 12 of the TIFF specification.

Table 1.9 describes the “Data Sample Format” defined in Section 19 of the TIFF specification.

Table 1.10 describes the “RGB Image Colorimetry” defined in Section 20 of the TIFF specification.

Table 1.11 describes the “YCbCr Images” defined in Section 21 of the TIFF specification.

Table 1.12 describes the “JPEG Compression” defined in Section 22 of the TIFF specification.

Legend for TIFF table:

- Columns Field, Description, Tag, and Type refer to corresponding specification items of tag according to

TIFF specifications.

- ROCN column specifies presence of the item:

o R: required o O: optional o C: conditional (Condition must be specified) o N: Do not use

- Restricted field values: indicates (when applicable) required values for TIFF tag.

Table 1.1: TIFF 6.0 Baseline Tags

TIFF 6.0

Section 8

Baseline Field Reference Guide

Field Description Tag Type ROCN Restricted Field Values

Artist Person who created the image.

315 ASCII R Populate with the name of the organization who created the file.

BitsPerSample Number of bits per component.

258 SHORT R…