Forums

Hi guys. Bit of a technical question here. I am experimenting with FMV on Arc and in QGIS. I used tutorial data to learn more about FMV, and then acquired UAS video from my employer.  These videos did not work on both platforms, and after investigation I found that this was due to there being 12 audio streams in the videos.  After getting rid of the audio using a video conversion application to get rid of the audio, I could sort of use it (In Arc I could see the video with the FMV plugin, but then realized there was no metadata in the converted video). I would like to know how I could convert video using COTS software, while keeping the metadata, but losing the audio, and how I could extract the metadata from the original video to save it as csv, and then recreate the video using multiplexer to combine the video and csv. Regards Pieter

This is a very interesting mapping platform for the agriculture community. The Belarus-based startup platform uses Sentinel-2 data and AI to instantly delineate thousands of crop fields and status of 20 plus crops in USA and Europe.  They also have smartphone-based apps which you can use to find these solutions for your field as well. The platform applies Machine Learning, which constantly improves the service as more data and feedback is collected. Considering that a mind-boggling 376,835,301 hectares of fields across Europe and the USA have already been analyzed and catalogued, the system has reached a remarkable level of maturity. OneSoil — a Copernicus-enabled start-up from Belarus Check out their interactive map. Onesoil homepage

Great, thanks for sharing  

Interesting articles :     North-South displacement field  - 1999 Hector-Mine earthquake, California   In complement to seismological records, the knowledge of the ruptured fault geometry and co-seismic ground displacements are key data to investigate the mechanics of seismic rupture. This information can be retrieved from sub-pixel correlation of optical images. We are investigating the use of SPOT (Satellite pour l'Observation de la Terre) satellites images. The technique developed here is attractive due to the operational status of a number of optical imaging programs and the availability of archived data. However, uncertainties on the imaging system itself and on its attitude dramatically limit its potential. We overcome these limitations by applying an iterative corrective process allowing for precise image registration that takes advantage of the availability of accurate Digital Elevation Models with global coverage (SRTM). This technique is thus a valuable complement to SAR interferometry which provides accurate measurements kilometers away from the fault but generally fails in the near-fault zone where the fringes get noisy and saturated. Comparison between the two methods is briefly discussed, with application on the 1992 Landers earthquake in California (Mw 7.3). Applications of this newly developped technique are presented: the horizontal co-seismic displacement fields induced by the 1999 Hector-Mine earthquake in California (Mw 7.1) and by the 1999 Chichi earthquake in Taiwan (Mw 7.5) have recently been retrieved using archive images. Data obtained can be downloaded (see further down) Latest Study Cases   Sub-pixel correlation of optical images   Following is the flow chart of the technique that as been developped. It allows for precise orthorectification and coregistration of the SPOT images. More details about the optimization process will be given in the next sections.             Understanding the disparities measured from Optical Images Differences in geometry between the two images to be registered:                 - Uncertainties on attitudes parameters (roll, pitch, yaw)              - Inaccuracy on orbital parameters (position, velocity)              - Incidence angle differences + topography uncertainties (parallax effect)              - Optical and Electronic biases (optical aberrations, CCD misalignment, focal length, sampling period, etc… )                        » May account for disparities up to 800 m on SPOT 1,2,3,4 images; 50m for SPOT 5 (see [3]). Ground deformations:                 - Earthquakes, land slides, etc…                         » Typically subpixel scale: ranging from 0 to 10 meters. Temporal decorrelation:                - Changes in vegetation, rivers, changes in urban areas, etc…                          » Correlation is lost: add noise to the measurement – up to 1m.             » Ground deformations are largely dominated by the geometrical artifacts.   Precise registration: geometrical corrections SPOT (Systeme pour l'Observation de la Terre) satellites are pushbroom imaging systems ([1],[2]): all optical parts remain fixed during acquisition and the scanning is accomplished by the forward motion of the spacecraft. Each line in the image is then acquired at a different time and submitted to the different variations of the platform. The orthorectification process consists in modeling and correcting these variations to produce cartographic distortion free images. It is then possible to accurately register images and look for their disparities using correlation techniques.                                           Attitude variations (roll, pitch, and yaw) during the scanning process have to be integrated in the image model (see [1],[2]). Errors in correcting the satellite look directions will result in projecting the image pixels at the wrong location on the ground: important parallax artifacts will be seen when measuring displacement between two images. Exact pixel projection on the ground is achieved through an optimization algorithm that iteratively corrects the look directions by selecting ground control points. An accurate topography model has to be used. What parameters to optimize?               - Initial attitudes values of the platform (roll, pitch, yaw),             - Constant drift of the attitude values along the image acquisition,             - Focal length (different value depending on the  instrument , HRG1 – HRG2),             - Position and velocity. How to optimize:   Iterative algorithm using a set of GCPs (Ground Control Points). GCPs are generated automatically with a subpixel accuracy: they result from a correlation between an orthorectified reference frame and the rectified image whose parameters are to be optimized. A two stages procedure: - One of the image is optimized with respect to the shaded DEM (GCP are generated from the correlation with the shaded DEM). The DEM is then considered as the ground truth. No GPS points are needed. - The other image is then optimized using another set of GCP resulting from the correlation with the first image (co-registration).   Measuring co-seismic deformation with InSAR, a comparison                                 A fringe represents a near-vertical displacement of 2.8 cm SAR interferogram (ERS): near-vertical component of the ground displacement induced by the 1992 Landers earthquake [Massonnet et al., 1993]. No organized fringes in a band within 5-10 km of the fault trace: displacement sufficiently large that the change in range across a radar pixel exceeds one fringe per pixel, coherence is lost. http://earth.esa.int/applications/data_util/ndis/equake/land2.htm      » SAR interferometry is not a suitable technique to measure near fault displacements The 1992 Landers earthquake revisited:                                                                                                                       Profile in offsets and elastic modeling show good agreement   From: [6] - Measuring earthqakes from optical satellite images, Van Puymbroeck, Michel, Binet, Avouac, Taboury - Applied Optics Vol. 39, No 20, 10 July 2000 Other applications of the technique, see [4], [5].      » Fault ruptures can be imaged from this technique   Applying the precise rectification algorithm + subpixel correlation: The 1999 Hector-Mine earthquake (Mw 7.1, California)   Obtaining the Data (available in ENVI file Format. Load banbs as gray scale images. Bands are: N/S offsets, E/W offsets, SNR): Raw and filtered results: HectorMine.zip Pre-earthquake image: SPOT 4, acquisition date: 08-17-1998 Ground resolution: 10m Post-earthquake image: SPOT 2, acquisition date: 08-18-2000 Ground resolution: 10m Offsets measured from correlation: Correspond to sub-pixel offsets in the raw images. Correlation windows: 32 x 32 pixels 96m between two measurements So far we have:                 - A precise mapping of the rupture zone: the offsets field have a resolution of 96 m,                 - Measurements with a subpixel accuracy (displacement of at most 10 meters),                 - Improved the global georeferencing of the images with no GPS measurements,                 - Improved the processing time since the GCP selection is automatic,                 - Suppressed the main attitude artifacts. The profiles do not show any long wavelength deformations (See Dominguez et al. 2003) We notice:                 - Linear artifacts in the along track direction due to CCD misalignments,                       Schematic of a DIVOLI showing four CCD linear arrays.                 - Some topographic artifacts: the image resolution is higher than the DEM one,              - Several decorrelations due to rivers and clouds,              - High frequency noise due to the noise sensitivity of the Fourier correlator (See Van Puymbroeck et al.).   Conclusion             Subpixel correlation technique has been improved to overcome most of its limitations:                     » Precise rectification and co-registration of the images,                     » No more topographic effects (depending on the DEM resolution),                     » No need for GPS points – independent and automatic algorithm,                     » Better spatial resolution (See Van Puymbroeck et al.)             To be improved:                     » Stripes due to the CCD’s misalignment,                     » high frequency noise from the correlator,                     » Process images with corrupted telemetry. » The subpixel correlation technique appears to be a valuable complement to SAR interferometry for ground deformation measurements. References: [1] SPOT 5 geometry handbook: ftp://ftp.spot.com/outgoing/SPOT_docs/geometry_handbook/S-NT-73-12-SI.pdf [2] SPOT User's Handbook Volume 1 - Reference Manual: ftp://ftp.spot.com/outgoing/SPOT_docs/SPOT_User's Handbook/SUHV1RM.PDF [3] SPOT 5 Technical Summary ftp://ftp.spot.com/outgoing/SPOT_docs/technical/spot5_tech_slides.ppt [4] Dominguez S., J.P. Avouac, R. Michel Horizontal co-seismic deformation of the 1999 Chi-Chi earthquake measured from SPOT satellite images: implications for the seismic cycle along the western foothills of Central Taiwan, J. Geophys. Res., 107, 10 1029/2001JB00482, 2003. [5] Michel, R. et J.P., Avouac, Deformation due to the 17 August Izmit earthquake measured from SPOT images, J. Geophys. Res., 107, 10 1029/2000JB000102, 2002. [6] Van Puymbroeck, N., Michel, R., Binet, R., Avouac, J.P. and Taboury, J. Measuring earthquakes from optical satellite images, Applied Optics Information Processing, 39, 23, 3486–3494, 2000. Publications: Leprince S., Barbot S., Ayoub F., Avouac, J.P. Automatic, Precise, Ortho-rectification and Co-registration for Satellite Image Correlation, Application to Seismotectonics. To be submitted. Conferences: F Levy, Y Hsu, M Simons, S Leprince, J Avouac. Distribution of coseismic slip for the 1999 ChiChi Taiwan earthquake: New data and implications of varying 3D fault geometry. AGU 2005 Fall meeting, San Francisco. M Taylor, S Leprince, J Avouac.  A Study of the 2002 Denali Co-seismic Displacement Using SPOT Horizontal Offsets, Field Measurements, and Aerial Photographs. AGU 2005 Fall meeting, San Francisco. Y Kuo, F Ayoub, J Avouac, S Leprince, Y Chen, J H Shyu, Y Kuo. Co-seismic Horizontal Ground Slips of 1999 Chi-Chi Earthquake (Mw 7.6) Deduced From Image-Comparison of Satellite SPOT and Aerial Photos. AGU 2005 Fall meeting, San Francisco.   source: http://www.tectonics.caltech.edu/geq/spot_coseis/

Thank you Lurker for your kind response. Actually, I would like to visualize animal tracking using moveVis tools by creating path animations from geo-location point data. Thank you so much for the example.