With the development of medical endoscopic equipment, minimally invasive surgery (MIS) has gradually become an important technical means in daily medical practice and has been widely used. Reconstructing three-dimensional images from endoscopic images that meet the requirements of surgical accuracy is a key step to assist surgeons in accurate surgery. In recent years, the emergence of surgical robots has also put forward new requirements for extended endoscopic vision. However, the illumination reflection model based on experience has the problem of limited expressiveness for the imaging process in the real environment. In this paper, an illumination re-flection model based on irradiance and micro surface theory is adopted, and it is applied to the shadow shape recovery algorithm. By comparing the results with the shadow shape recovery al-gorithm based on empirical reflection model, it is shown that the illumination reflection model based on irradiance and micro surface theory adopted in this paper can more effectively recon-struct the three-dimensional information of the target surface, especially the reconstruction of fine texture features.