Multi-sensor 3D image fusion and interactive search

Ross, William N.; Waxman, Allen M.; Streilein, William W.; Aguiiar, M.; Verly, Jacques; Liu, F.; Braun, Michael; Harmon, P.; Rak, Steven J.

doi:10.1109/ific.2000.862685

Cited by 26 publications

(11 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The system is based on the proposed approach in [10] and later implemented in the context of multi-sensor fusion in [11], [12] and [13]. This is accomplished by allowing the user to define features of interest by selecting representative pixels on the imagery.…”

Section: Fusion-based Pattern Recognitionmentioning

confidence: 99%

“…By doing this, the learning system can leverage off the expert to solve the specific task of identifying the sampled structure across all available modalities. As in [11]- [13], this information is used to train an ART-based supervised neural network [14] known as Fuzzy ARTMAP [15] to recognize or segment the image based on the user's input. In addition, the resulting train network can be encapsulated and saved as an agent which can later be loaded to pre-screen images by highlighting areas of potential interest for the user.…”

Section: Fusion-based Pattern Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

Advances in the use of neurophysiologycally-based fusion for visualization and pattern recognition of medical imagery

Aguilar

Im²,

Hasanbelliu³

2003

Sixth International Conference of Information Fusion, 2003. Proceedings of The

View full text Add to dashboard Cite

show abstract

Section: Fusion-based Pattern Recognitionmentioning

confidence: 99%

Section: Fusion-based Pattern Recognitionmentioning

confidence: 99%

Advances in the use of neurophysiologycally-based fusion for visualization and pattern recognition of medical imagery

Aguilar

Im²,

Hasanbelliu³

2003

Sixth International Conference of Information Fusion, 2003. Proceedings of The

View full text Add to dashboard Cite

show abstract

“…Use of this form of learning has also been applied successfully to a variety of image mining and tracking tasks [2][3][4][5][6][7]. The speed and performance of this learning algorithm makes it suitable for real-time and interactive situations wherein an operator/analyst can help teach the model via simple point and click actions.…”

Section: Learning-based Pattern Recognitionmentioning

confidence: 99%

SeeCoast port surveillance

et al. 2006

View full text Add to dashboard Cite

SeeCoast extends the US Coast Guard Port Security and Monitoring system by adding capabilities to detect, classify, and track vessels using electro-optic and infrared cameras, and also uses learned normalcy models of vessel activities in order to generate alert cues for the watch-standers when anomalous behaviors occur. SeeCoast fuses the video data with radar detections and Automatic Identification System (AIS) transponder data in order to generate composite fused tracks for vessels approaching the port, as well as for vessels already in the port. Then, SeeCoast applies rule-based and learning-based pattern recognition algorithms to alert the watch-standers to unsafe, illegal, threatening, and other anomalous vessel activities. The prototype SeeCoast system has been deployed to Coast Guard sites in Virginia. This paper provides an overview of the system and outlines the lessons learned to date in applying data fusion and automated pattern recognition technology to the port security domain.Keywords: Port security, video surveillance, track fusion, alerts, anomaly detection, homeland security GOALS AND ARCHITECTUREAutomated scene understanding is a common issue throughout the Department of Homeland Security, including Coast Guard, Border Patrol, Transportation Security, and Infrastructure Protection. Numerous sensors have been deployed for these Agencies, engendering watch-stander manning as a growing problem. The goal of automated scene understanding is to reduce operator workload while maintaining effectiveness and improving detection of anomalous behaviors.SeeCoast builds upon the USCG Hawkeye system, which uses a combination of surface surveillance radar and Automatic Identification System (AIS) beacons to track vessels and generate zone-incursion alerts. Hawkeye also includes manually-controlled and radar-slaved video cameras, an integrated command center, and a web-based information client. Hawkeye has been deployed to more than six operational US Coast Guard (USCG) Sector Command Centers. It allows watch-standers to surveil all vessel arrivals, departures, and transits within a port-centered area of interest.Currently, watch-standers attempt to detect and assess possible anomalous behaviors using a large number of video cameras, radars, and AIS track data. SeeCoast has two goals: to increase the effectiveness of the watch-standers, and to increase their efficiency in surveillance operations. SeeCoast increases watch-stander effectiveness by adding capabilities to process video and infrared camera streams to detect, classify (small, medium, large vessel size), and track vessels; to fuse video tracks with radar & AIS tracks; to maintain seamless 24/7 vigilance on the track scene; to provide consistent alerting using established rules and learned normalcy models; and to free the surveillance system operators to assess behaviors rather than spending their time detecting behaviors. SeeCoast increases the efficiency of maritime surveillance operations by simultaneously examining data and tracks from many...

show abstract

“…The resulting diagram of the relationships among classes can then guide the construction of consistent layered maps. Inputs for the Boston testbed example shown in Figure 4 were preprocessed by a version of the Lincoln Lab image mining system [15][16][17] [29,30]. For each pixel in the Boston image, this Module, implemented on an ERDAS Imagine (gis.lcica-geosystems.com) platform, produced a 41-dimensional input vector representing local contrast, color, and texture attributes.…”

Section: Deriving Consistent Knowledge From Inconsistent Infor Mationmentioning

confidence: 99%

“…Claimed savings in manufacturing costs are in millions of dollars per annum." At Lincoln Lab, a team led by Waxman developed an image mining system which incorporates several BU/CNS-bascd models of vision and recognition [15][16][17]. Over the years a dozen CNS graduates have contributed to this effort, which is now located at Alphatcch, Inc.…”

Section: Introductionmentioning

confidence: 99%