CEANOGRAPHERS have long been imaging the water column and sea floor to aid in the study, analysis, and interpretation of biological, physical oceanographic, geological, and archaeological data. However, the underwater environment presents a unique set of constraints that have limited our ability to collect and process underwater imagery. Some of the fundamental constraints in imaging underwater are associated with the physics of the optical sensing modality. The rapid attenuation of the electromagnetic spectrum underwater, for instance, prevents the user from framing large objects within the field of view of a single frame, thus making it difficult to obtain a global perspective of sites of interest. The effect of backscatter from suspended particles in the water column is also a major factor in limiting our ability to image underwater. System issues can also dominate the imaging application. Vehicle dynamics on underwater platforms make it difficult, for example, to focus on and study individual organisms in the mid-water column. Navigation issues can also alias efforts to detect change in the environment as sensors may deliver information at resolutions that are far greater than our ability to place them in the real world. Some of the research issues underwater have terrestrial analogs, such as the tasks associated with reconstructing three-dimensional information about the world from its two-dimensional projections. But the added complexity of working in unstructured environments underwater in the presence of moving light sources makes these tasks especially challenging. Finally, one must also consider the case where the sensors are too successful in gathering data and researchers must face the daunting task of wading through hours of video tape or thousands of images to manage, index and extract useful information from a deluge of imagery. This issue of the IEEE JOURNAL OFOCEANIC ENGINEERINGis devoted to the topic of image and video processing. This seems particularly useful at this juncture, as a confluence of technologies now allow us to examine and explore the full potential associated with effectively utilizing imaging sensors underwater. The exponential rise in computing and graphics; the availability of high-resolution digital cameras, multispectral sensors, hyperspectral sensors, and acoustic imaging systems; and the variety of underwater platforms with associated navigation and inertial sensors have all contributed to pushing the limits of image-processing applications. What is exciting about the papers included in this special issue is that they are fairly representative of the entire gamut of applications and methodologies.