Traditional remote sensing imaging for marine oil spills faces challenges such as high sensitivity to lighting conditions, low oil?water contrast, and significant surface glare interference. To address the resulting degradation in imaging quality and reduced accuracy of monitoring systems, this study employs four imaging modalities—visible light, visible light polarization, near-infrared, and long-wave infrared—to conduct outdoor field experiments under three high-interference scenarios (strong light, weak light, and sun glint conditions) from the perspective of oil spill imaging channel selection. By comparatively analyzing the information entropy and strong edge width of oil spill images across the four imaging modes, the study verifies the feasibility of using visible light polarization imaging under strong/weak light and sun glint conditions, and near-infrared imaging under strong light and sun glint conditions for oil spill detection. Meanwhile, experiments demonstrate that long-wave infrared exhibits limited effectiveness for thin oil film detection under weak illumination or sun glint conditions. This research provides experimental evidence for selecting optimal information channels in marine oil spill monitoring systems.