问题描述：
英语翻译
Data analysis begins with digital videotapes recorded at the LMT being sent to JSC for processing.Each
compressed video tape is loaded from a SONY DHR-1000 DV recorder/player onto a dual-processor 500 MHz
Pentium III PC through an IEEE 1394 PC1 board.This transfer is effected digitally without loss of information and
is effected in real-time,i.e.it takes one hour to transfer one hour of digital video from magnetic tape to a computer
file.The transfer is controlled via a dialog-based software application written in Visual C++.One hour of DV
compressed video stripped of audio comprises about 12 GB of storage,so an 18 GB drive is sufficient to hold both
an hour of video plus the output from the detection software.
Once the digital videotape is loaded onto PC,automated debris detection software is applied.This software
begins by decoding 200 compressed DV video images.Each video image pixel consists of three l-byte values; one
each for red,green,and blue.A weighted sum of these three values yields a single intensity value for each pixel.
The DV format compresses RGB information as fixed 120,000 bytes per image frame (Sony 1996).Each image
frame is composed of two fields; an even field containing the even numbered horizontal lines and an odd field.
Fields are recorded at 59.94 fields per second.Two preliminary subroutines are applied to the 200 decoded images
to determine the indices of image pixels within the camera FOV,the rate and direction through the image at which
the star field crosses the FOV,and a camera rotation angle.Then the main debris detection subroutine begins.This
subroutine runs at 12 times real time,taking 12 hours of computer time to detect the debris,meteors,and satellites
in one hour of digital video.This automated debris detection and tracking algorithm consists of the following
steps.
1) Decode the current color image frame and map it onto 640x480 pixel 1:1 aspect ratio images.
2) Form an estimate of the star field plus sky-background and subtract it from each image
3) Apply a library of matched filters to each image.
4) Find the maximum matched filter output for each of seven altitude bins.
5) Update a sample mean and variance for the maximum matched filter output in each of seven altitude bins.
6) Apply a dual-statistic hypothesis test to determine whether a debris object is present.The two statistics are
statistically independent,increasing the confidence level for the combined two-statistic test.
7) Track the detected debris objects through the FOV.Perform a quantitative check on the tracked history of
each object to further reduce false detections.
The automated detection software stores video clips of each detected satellite,debris,and meteor and providesestimates of the,altitude and inclination of the orbits.Once this processing is complete,the results are reviewed
and analyzed using a semi-automated Event-Reviewer software.
Data analysis begins with digital videotapes recorded at the LMT being sent to JSC for processing.Each
compressed video tape is loaded from a SONY DHR-1000 DV recorder/player onto a dual-processor 500 MHz
Pentium III PC through an IEEE 1394 PC1 board.This transfer is effected digitally without loss of information and
is effected in real-time,i.e.it takes one hour to transfer one hour of digital video from magnetic tape to a computer
file.The transfer is controlled via a dialog-based software application written in Visual C++.One hour of DV
compressed video stripped of audio comprises about 12 GB of storage,so an 18 GB drive is sufficient to hold both
an hour of video plus the output from the detection software.
Once the digital videotape is loaded onto PC,automated debris detection software is applied.This software
begins by decoding 200 compressed DV video images.Each video image pixel consists of three l-byte values; one
each for red,green,and blue.A weighted sum of these three values yields a single intensity value for each pixel.
The DV format compresses RGB information as fixed 120,000 bytes per image frame (Sony 1996).Each image
frame is composed of two fields; an even field containing the even numbered horizontal lines and an odd field.
Fields are recorded at 59.94 fields per second.Two preliminary subroutines are applied to the 200 decoded images
to determine the indices of image pixels within the camera FOV,the rate and direction through the image at which
the star field crosses the FOV,and a camera rotation angle.Then the main debris detection subroutine begins.This
subroutine runs at 12 times real time,taking 12 hours of computer time to detect the debris,meteors,and satellites
in one hour of digital video.This automated debris detection and tracking algorithm consists of the following
steps.
1) Decode the current color image frame and map it onto 640x480 pixel 1:1 aspect ratio images.
2) Form an estimate of the star field plus sky-background and subtract it from each image
3) Apply a library of matched filters to each image.
4) Find the maximum matched filter output for each of seven altitude bins.
5) Update a sample mean and variance for the maximum matched filter output in each of seven altitude bins.
6) Apply a dual-statistic hypothesis test to determine whether a debris object is present.The two statistics are
statistically independent,increasing the confidence level for the combined two-statistic test.
7) Track the detected debris objects through the FOV.Perform a quantitative check on the tracked history of
each object to further reduce false detections.
The automated detection software stores video clips of each detected satellite,debris,and meteor and providesestimates of the,altitude and inclination of the orbits.Once this processing is complete,the results are reviewed
and analyzed using a semi-automated Event-Reviewer software.
问题解答：
我来补答展开全文阅读