How does the TOMHT handles the input of detectable branchID's?
2 ビュー (過去 30 日間)
古いコメントを表示
Hi again,
First of all thanks for all the support I have gotten so far. I have another question. The tomht tracker provides the option to pass on a list of branch ID's that will be considered detectable in the next update. I was wondering, how does the TOMHT tracker handle this input? Does it simply predicts these branches and handles the score differently for these branches?
Kind regards,
Joost
0 件のコメント
採用された回答
Elad Kivelevitch
2022 年 3 月 30 日
Hi Joost,
Happy to help.
For your question: the way the trackerTOMHT uses the detectableBranchIDs is that it uses the probability of detection of the detectable branches (or, if not provided in the second column of the input, the tracker PD) when calculating their score after a miss. For undetectable branches, PD is zero, so the score does not decrease. This allows you to keep undetectable or poorly detectable branches for a larger number of updates relative to not using that.
The tracker does not use it in the association. So, if a certain branch ID is considered undetected or poorly detected, but is assigned to a detection after all, the tracker allows the assignment to happen.
HTH,
Elad
その他の回答 (0 件)
参考
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!