Visual localization algorithm based on binocular matching in seedling robots
When intelligent agricultural equipment is used in agricultural production,the precise positioning of crops can provide decision-making for seedling work.In order to solve the problem of the accuracy of binocular positioning of seedling stems,an improved SURF algorithm was proposed to complete the measurement of the three-dimensional coordinates of seedling stems.Firstly,according to the color difference between the crop and the soil background,the super-green algorithm was used to segment the crop outline,and the improved SURF algorithm was used to complete the left and right image matching of binocular vision,so that the three-dimensional coordinates of the seedling stem could be determined according to the coordinate conversion relationship.The results showed that the average absolute error in the X direction was 3.5232 mm,that the average absolute error in the Y direction was 4.194 mm,and that the average absolute error in the Z direction was 3.045 mm.The success rate of identification met the expected requirements,which effectively improved the measurement accuracy of seedling stems.
万方数据
维普
