TECHNOLOGY AREA(S): Battlespace, Electronics, Sensors

ACQUISITION PROGRAM: Coastal Battlefield Reconnaissance and Analysis (COBRA), PMS 495, Mine Warfare Program Office

The technology within this topic is restricted under the International Traffic in Arms Regulation (ITAR), 22 CFR Parts 120-130, which controls the export and import of defense-related material and services, including export of sensitive technical data, or the Export Administration Regulation (EAR), 15 CFR Parts 730-774, which controls dual use items. Offerors must disclose any proposed use of foreign nationals (FNs), their country(ies) of origin, the type of visa or work permit possessed, and the statement of work (SOW) tasks intended for accomplishment by the FN(s) in accordance with section 3.5 of the Announcement. Offerors are advised foreign nationals proposed to perform on this topic may be restricted due to the technical data under US Export Control Laws.

OBJECTIVE: Develop an automated curvilinear mineline detection algorithm.

DESCRIPTION: The Navy is interested in technologies that facilitate automated target pattern recognition capabilities in aerial multi-spectral images of curvilinear-arranged targets in various coastal and inland environments. COBRA multi-spectral imagery has wavelengths in the visible spectrum, including those that penetrate the water, and infrared. These minefields may be placed in a variety of configurations. Current patterned algorithms exploit a mineline’s linear features and rely on minimal variations in mine-to-mine placement angles to make a detection. This tradeoff improves the algorithm’s linear mineline performance. Automated target recognition algorithms that annotate minefields placed in nonlinear patterns would reduce mission execution time during the post mission analysis phase and improve detection system performance.

Typically, minelines placed in a coastal environment follow the natural landforms of the area and may take on complex, non-linear shapes. Accurate and reliable automatic detection and notification of the presence of these curvilinear minelines would reduce operator review time to mark the area for clearance or avoidance by follow-on forces. Studies have shown that an accurate and reliable automatic detection algorithm reduced detection time and improved detection rate. If all of the algorithm’s cues are false alarms, operator performance may be worse than if no aiding was provided at all. This would reduce the mission time and the potential for error due to operator fatigue and human error.

The Navy needs innovative methods that can recognize non-linear, patterned targets in a variety of inland and coastal environments as imaged aerially with a multi-spectral camera. The proposed effort will develop algorithms for automated target recognition of curvilinear minelines to optimize Probability of Detection (PD) and Probability of False Alarm (PFA)/False Alarm Rate (FAR) of the COBRA Block I System. Targets will have a top surface area equivalent to that of a circle with a diameter of approximately 15 to 30 centimeters, which equates to approximately 6-14 pixels in COBRA’s imagery. In order to work within the current COBRA Block I Real Time Processor (RTP) framework, the algorithms will need to be modular as the RTP uses independent algorithm libraries. Modules will perform logically discrete functions and provide well-defined interfaces for other modules. Algorithms will be hardware agnostic, but for development considerations only, will run on an Intel-based 64-bit architecture system with discrete NVIDIA graphics cards. As newer hardware becomes available, the algorithm kernels should be capable of scaling to utilize available resources. These modular algorithms will be integrated into the COBRA Airborne Payload Subsystem (CAPS), the COBRA Post Mission Analysis (PMA) Subsystem, and potentially other flight and post-mission analysis systems as identified. The algorithms will be implemented as object-oriented C++ for Central Processing Units (CPUs) and/or Open Computing Language (OpenCL) or Compute Unified Device Architecture (CUDA) for Graphics Processing Unit (GPU) processing. Processing techniques should work in conjunction with the current RTP framework and algorithms to process imagery in real time; currently an image to be processed is captured every 763 milliseconds. The proposed algorithms will be required to conform to the Navy’s Open Architecture (OA) initiative. Modular design of software components will enable openness to the Navy and others.

The Phase II effort will likely require secure access, and NAVSEA will process the DD254 to support the contractor for personnel and facility certification for secure access. The Phase I effort will not require access to classified information. If need be, data of the same level of complexity as secured data will be provided to support Phase I work.

Work produced in Phase II may become classified. Note: The prospective contractor(s) must be U.S. Owned and Operated with no Foreign Influence as defined by DoD 5220.22-M, National Industrial Security Program Operating Manual, unless acceptable mitigating procedures can and have been be implemented and approved by the Defense Security Service (DSS). The selected contractor and/or subcontractor must be able to acquire and maintain a secret level facility and Personnel Security Clearances, in order to perform on advanced phases of this contract as set forth by DSS and NAVSEA in order to gain access to classified information pertaining to the national defense of the United States and its allies; this will be an inherent requirement. The selected company will be required to safeguard classified material IAW DoD 5220.22-M during the advance phases of this contract.

PHASE I: Develop a concept for an algorithm capable of detecting curvilinear minelines in a variety of inland and coastal environments using aerial multispectral imagery. Demonstrate the feasibility of the algorithm through modeling and simulation. Develop a Phase II plan. The Phase I Option, if exercised, will include the initial design specifications and capabilities description to build a prototype solution in Phase II.

PHASE II: Develop and deliver a modular software library prototype to provide efficient real-time detection of curvilinear minelines using COBRA Block I imagery as described in the Description. The prototype may run in a development environment that meets the hardware performance specifications and software libraries of the COBRA Block I RTP. Generate a performance estimation of the developed capability to include PD, PFA/FAR, operating time, and operational impacts of environmental conditions including clutter and vegetation. Use operationally representative data for the evaluation. Ensure that the algorithm performance meets the system’s minefield detection performance using specified target sizes. Prepare a Phase III development plan to transition the technology for Navy and potential commercial use.

It is probable that the work under this effort will be classified under Phase II (see Description section for details).

PHASE III DUAL USE APPLICATIONS: Support the Navy in transitioning the technology for Navy use. While algorithm modularity eases integration, integrate the algorithms into the RTP. Perform the following integration tasks: adding the algorithms into the existing processing framework, load balancing across the RTP’s various processors, and acceptance testing in the operational configuration. Further refine the software to ensure compatibility with existing mine warfare operator interfaces and workstations according to the Phase III SOW. Support updates to the COBRA Technical Data Package to support the Navy in transitioning the design and technology into the COBRA Production baseline for future Navy use.

The technology developed here can be applied to pattern recognition problems, surveillance tasks, remote sensing, and Intelligence Preparation of the Operational Environment (IPOE). Commercial applications include biometrics, computer vision, facial recognition, and histopathology.

REFERENCES:

1. "AN/DVS-1 Coastal Battlefield Reconnaissance and Analysis (COBRA)." The U.S. Navy – Fact File. Last update 4 October 2017. http://www.navy.mil/navydata/fact_display.asp?cid=2100&tid=1237&ct=2

2. Bernabe, Sergio, Lopez, Sebastian, Plaza, Antonio, and Sarmiento, Roberto. “GPU Implementation of an Automatic Target Detection and Classification Algorithm for Hyperspectral Image Analysis.” IEEE Geoscience and Remote Sensing Letters, Vol. 10, No. 2, March 2013. https://ieeexplore.ieee.org/abstract/document/6218752/

3. Reiner, Adam J., Hollands, Justin G., and Jamieson, Greg A. “Target Detection and Identification Performance Using an Automatic Target Detection System.” Human Factors, Vol. 59, No. 2, 01 March 2017, pp. 242-258. https://doi.org/10.1177/0018720816670768

4. Samson, Joseph W., Witter, Lester J., Kenton, Arthur C., and Holloway, John H. “Real-time Implementation of a Multispectral Target Detection Algorithm.” SPIE 5089, Detection and Remediation Technologies for Mines and Minelike Targets VIII, 11 September 2003. https://doi.org/10.1117/12.501567

5. El-Saba, Aed, Alam, Mohammad S., and Sakla, Wesam A. “Pattern Recognition via Multispectral, Hyperspectral, and Polarization-based Imaging.” SPIE Defense, Security and Sensing, Proceedings Volume 7696, Automatic Target Recognition XX; Acquisition Tracking, Pointing, and Laser System Technologies XXIV; and Optical Pattern Recognition XXI, 13 May 2010. https://doi.org/10.1117/12.851689

KEYWORDS: Automated Target Detection; Automated Pattern Detection; Curvilinear Minefields; Coastal Battlefield Reconnaissance and Analysis; COBRA; Post Mission Analysis; Mine Countermeasures; MCM