Pavement condition rating method
By using a plurality of images obtained by combining image of damage included in a road surface still image based on perspective and definitions of the damage as damage learning data, an object detection model is built. A plurality of data wherein the number of rectangles in an analysis result image displaying the rectangles surrounding the damage output as an analysis result of the object detection model and a pavement condition rating result obtained by visual inspection of the analysis result image of the object detection model are combined is used as condition rate learning data to build a rating estimation model, and pavement condition is rated using the rating estimation model.
1. A pavement condition rating method wherein:
a plurality of images obtained by combining image of damage included in a road surface still image based on perspective and definitions of the damage are used as damage learning data,
an object detection model for performing image analysis wherein an image wherein an object detected in an analysis target image is surrounded by a rectangle is output as an analysis result is built using the damage learning data,
damages included in a road surface still image based on perspective are detected using the object detection model,
a plurality of data wherein the number of rectangles in an analysis result image displaying the rectangles surrounding the damage output as an analysis result and a pavement condition rating result obtained by visual inspection of the analysis result image are combined is used as condition rate learning data to build a rating estimation model,
and pavement condition is rated using the rating estimation model, wherein the damage learning data comprises a plurality of types of the damage, and a plurality of the analysis result images used for obtaining the pavement condition rating result and the number of the rectangles composing the condition rate learning data have a standard deviation of a sum of the number of the rectangles for each type of the damage in each of the analysis result images exceeding 20.