The detection of anomalous heat source point targets in a wide range of infrared systems requires a balance between the pixel resolution and temperature sensitivity. Given that the scale of the detector is stable, the existing space-borne infrared payloads have insufficient sensitivity when the amplitude is large and small amplitude when the spatial resolution is high. In response to these problems, in this study, we propose a preliminary plan for detecting heat sources at abnormal points in a large field of view using a time-delay integration (TDI) algorithm to process images. Under certain conditions, the temperature sensitivity requirements for detection can be satisfied and a theoretical width of 217 km × 122 km can be achieved. A set of high-sensitivity infrared imaging experimental systems was built, and testing and simulation-based detection experiments were conducted. The results show that the sensitivity performance of this scheme is approximately 37 mK when a width of 202 km×.114 km is realized, which fulfills the requirements for detecting a large-scale abnormal point heat source. Considering the solar reflectance of the target and background, a pixel resolution of 200 m is used in practical applications, which corresponds to a width of 128 km×102 km.