Image processing device, image processing method, and storage medium

1 . An image processing device comprising:

a storage medium configured to store an instruction readable by a computer and a hardware processor that is connected to the storage medium,

wherein the hardware processor executes the instruction readable by the computer, thereby

setting an initial value of at least one parameter of a group of initial parameters associated with a function that approximates a road boundary in an image representing an area in front of a mobile object based on a scaling function that gives a straight line passing through an identification value associated with an other parameter of the group of initial parameters and a vanishing point of the image, wherein the image is captured by a camera mounted on the mobile object;

iteratively updating, as a group of updated parameters, the group of initial parameters at a predetermined time point based on a group of previous parameters at a previous time point to the predetermined time point and constraint conditions set for the group of previous parameters; and

executing driving control or driving assistance of the mobile object based on the road boundary approximated by the function that is associated with the group of updated parameters, wherein the hardware processor iteratively updates the group of initial parameters associated with the function that approximates the road boundary by dividing the image at predetermined intervals into areas in a vehicle travel direction, iteratively updating the group of initial parameters for each element function that approximates the road boundary for each divided area, and taking a sum of the each element function.

2 . The image processing device according to claim 1 ,

wherein the function is a quadratic function, the hardware processor sets a linear coefficient initial value of a linear coefficient of the quadratic function on the basis of the identification value of the road boundary recognized in the lateral direction of the image, and the hardware processor iteratively updates the parameters at the predetermined time point by setting forgetting gains of secondary coefficients and linear coefficients of the group of previous parameters at the previous time point to the predetermined time point to values smaller than 1.

3 . The image processing device according to claim 1 ,

wherein the hardware processor uses a learned model learned to output a probability value indicating an existence probability of a road boundary for each of coordinates of the image in response to an input of the image to acquire the probability value and update the parameters on the basis of the probability value and the coordinates.

4 . The image processing device according to claim 3 ,

wherein the hardware processor iteratively updates the group of initial parameters to minimize an error between a component of the coordinates in a first direction and an estimated value of the component in the first direction calculated on the basis of a component of the coordinates in a second direction.

5 . An image processing method comprising:

by a computer,

setting an initial value of at least one parameter of a group of initial parameters associated with a function that approximates a road boundary in an image representing an area in front of a mobile object based on a scaling function that gives a straight line passing through an identification value associated with an other parameter of the group of initial parameters and a vanishing point of the image, wherein the image is captured by a camera mounted on the mobile object;

iteratively updating, as a group of updated parameters, the group of initial parameters at a predetermined time point based on a group of previous parameters at a previous time point to the predetermined time point and constraint conditions set for the group of previous parameters; and

executing driving control or driving assistance of the mobile object on the basis of the road boundary approximated by the function that is associated with the group of updated parameters, wherein the group of initial parameters associated with the function that approximates the road boundary is iteratively updated by dividing the image at predetermined intervals into areas in a vehicle travel direction, and wherein the group of initial parameters for each element function that approximates the road boundary for each divided area is iteratively updated, and a sum of the each element function is determined.

6 . A computer-readable non-transitory storage medium that stores a program causing a computer to execute:

setting an initial value of at least one parameter of a group of initial parameters associated with a function that approximates a road boundary in an image representing an area in front of a mobile object based on a scaling function that gives a straight line passing through an identification value associated with an other parameter of the group of initial parameters and a vanishing point of the image, wherein the image is captured by a camera mounted on the mobile object,

iteratively updating, as a group of updated parameters, the group of initial parameters at a predetermined time point based on a group of previous parameters at a previous time point to the predetermined time point and constraint conditions set for the group of previous parameters, and

executing driving control or driving assistance of the mobile object based on the road boundary approximated by the function that is associated with the group of updated parameters, wherein the group of initial parameters associated with the function that approximates the road boundary is iteratively updated by dividing the image at predetermined intervals into areas in a vehicle travel direction, and wherein the group of initial parameters for each element function that approximates the road boundary for each divided area is iteratively updated, and a sum of the each element function is determined.

7 . An image processing device comprising:

a storage medium configured to store an instruction readable by a computer and a processor that is connected to the storage medium,

wherein the processor executes the instruction readable by the computer, thereby

setting an initial value of at least one of parameters of a function that approximates a road boundary in an image representing an area in front of a mobile object, captured by a camera mounted on the mobile object, on the basis of identification values of other parameters among the parameters;

iteratively updating the parameters at a predetermined time point on the basis of the parameters at a previous time point to the predetermined time point and constraint conditions set for the parameters; and

executing driving control or driving assistance of the mobile object on the basis of the road boundary approximated by the function that has the updated parameters,

wherein the processor iteratively updates parameters of a function that approximates a road boundary by dividing the image at predetermined intervals into areas in a vehicle travel direction, iteratively updating the parameters of each element function that approximates the road boundary for each divided area, and taking a sum of the element functions,

wherein the function is a quadratic function,

the processor sets an initial value of a linear coefficient of the quadratic function on the basis of an identification value of the road boundary recognized in a lateral direction of the image,

the processor sets different forgetting gains according to degrees of coefficients of the quadratic function,

and the processor iteratively updates the parameters at a predetermined time point by setting forgetting gains of secondary coefficients and the linear coefficients of the parameters at the previous time point to the predetermined time point to values smaller than 1.