TY - GEN
T1 - Integration of Biologically Inspired Pixel Saliency Estimation and IPDA Filters for Multi-target Tracking
AU - Griffiths, Daniel
AU - Badriasl, Laleh
AU - Scoleri, Tony
AU - Brinkworth, Russell S.A.
AU - Arulampalam, Sanjeev
AU - Finn, Anthony
PY - 2020
Y1 - 2020
N2 - The ability to visually locate and maintain targets is fundamental to modern autonomous systems. By augmenting a novel biologically-inspired target saliency estimator with Integrated Probability Data Association (IPDA) filters and linear prediction techniques, this paper demonstrates the reliable detection and tracking of small and weak-signature targets in cluttered environments. The saliency estimator performs an adaptive, spatio-temporal tone mapping and directional filtering that enhances local contrast and extracts consistent motion, strengthened by the IPDA mechanism which incrementally confirms targets and further removes stochastic false alarms. This joint technique was applied to mid-wave infra-red imagery of maritime scenes where heavy sea clutter distracts significantly from true vessels. Once initialised, the proposed method is shown to successfully maintain tracks of dim targets as small as 2×1 pixels with 100% accuracy and zero false positives. On average, this method scored a sensitivity of 93.2%, with 100% precision, which surpasses well-established techniques in the literature. These results are very encouraging, especially for applications that require no misses in highly cluttered environments.
AB - The ability to visually locate and maintain targets is fundamental to modern autonomous systems. By augmenting a novel biologically-inspired target saliency estimator with Integrated Probability Data Association (IPDA) filters and linear prediction techniques, this paper demonstrates the reliable detection and tracking of small and weak-signature targets in cluttered environments. The saliency estimator performs an adaptive, spatio-temporal tone mapping and directional filtering that enhances local contrast and extracts consistent motion, strengthened by the IPDA mechanism which incrementally confirms targets and further removes stochastic false alarms. This joint technique was applied to mid-wave infra-red imagery of maritime scenes where heavy sea clutter distracts significantly from true vessels. Once initialised, the proposed method is shown to successfully maintain tracks of dim targets as small as 2×1 pixels with 100% accuracy and zero false positives. On average, this method scored a sensitivity of 93.2%, with 100% precision, which surpasses well-established techniques in the literature. These results are very encouraging, especially for applications that require no misses in highly cluttered environments.
KW - Infra-red
KW - Small target detection
KW - Target tracking
KW - Tone mapping
KW - Visual saliency
KW - Weak signature
UR - http://www.scopus.com/inward/record.url?scp=85081604481&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-41404-7_54
DO - 10.1007/978-3-030-41404-7_54
M3 - Conference contribution
AN - SCOPUS:85081604481
SN - 9783030414030
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 764
EP - 778
BT - Pattern Recognition - 5th Asian Conference, ACPR 2019, Revised Selected Papers
A2 - Palaiahnakote, Shivakumara
A2 - Sanniti di Baja, Gabriella
A2 - Wang, Liang
A2 - Yan, Wei Qi
PB - Springer
CY - Switzerland
T2 - 5th Asian Conference on Pattern Recognition, ACPR 2019
Y2 - 26 November 2019 through 29 November 2019
ER -