A passive underwater imaging system has been developed using a volume array of only 6 elements and time average product (TAP) broadband array processing techniques to create images consisting of thousands of pixels. This system was developed to image objects using ambient noise as illumination. Sea trials have been conducted where targets were passively detected at ranges up to 40 m. The system has also been used to image the global directivity of ambient noise, identifying the dominant noise sources, and to visually track the bearing and range of boat traffic and submerged divers. This system is capable of generating a 360° image of the ambient noise field approximately every second (with 1.6° resolution at 25-kHz bandwidth). The small number of transducers and relatively low data volume produced in the image processing make this system ideal for long-term monitoring of ocean noise. An autonomous version has been designed for offshore deployment. Initially it will consist of an autonomous stationary platform but will subsequently be incorporated into an autonomous underwater vehicle (AUV) for mobile noise surveillance missions. [Work supported by ONR.]