Historical building identification and detection method based on deep learning and high-resolution images

Embodiments of the present application disclose a method, apparatus, device and storage medium for recognizing and detecting historical buildings based on deep learning and high-resolution images, belonging to the field of object recognition technology, the method comprising: acquiring a remotely sensed building picture, dividing the building picture into a number of cell grids; judging four boundary boxes of the cell grids, judging whether the four boundary boxes are building boundary boxes; acquiring a cell building cells in the grid and remove the bounding boxes to obtain the building construct and obtain the predicted feature value of the building construct; based on the pre-trained building category training set, judge whether the predicted feature value of the building in the picture satisfies the predicted feature value threshold of the building classification in the training set; if it does, obtain the category information corresponding to the building classification, determine it as the category of the building in the picture, and complete the recognition. The present application helps to improve the accuracy and robustness when recognizing pictures and facilitates the investigation work of historical buildings. O bti reotl -olete buldn imgs divid sai buildings S IS I Judin th I fou bondn b Sxe s I *h cel grid jugn whte th orbudn boxe ar I buldn bondn boes if thyaennbidn-onigbxsetn them~~~ ~ ~ tS 0, Itewie *f the ar.uligbudigbxs0 etn te o1ugn whthrth -ISvlu f h fou bondn boxe If * h celS gri *s - iftear nonobtinng hecel igre2nigt henn0vle