Abstract: This study summarizes the methods and study progress of the collection of stumpage factors in forest resources inventory,probes into the new techniques and methods of collecting stumpage factors, and puts forward that the traditional collection methods are tedious, require a lot of manpower and material resources, and the accuracy of the results is not high. On the other hand, emerging technologies and equipment are usually expensive, not easy to carry, and data processing is complex and difficult. It is revealed that the main problems in the study of vision sensor technology are that there is no own image data set, the generalization ability and accuracy of the model are not high, and the occlusion problem of forest community sample trees is difficult to solve. In view of the existing problems, it is proposed to strengthen the study on new technologies and methods, especially in the field of visual sensors, it is necessary to reasonably construct forest community image data sets and improve the model generalization ability and accuracy of visual sensors in stumpage measurement, use the multi-sensor fusion technology to crack the occlusion.