Object detection device and object detection computer program

An object detection device, comprising:

• a memory configured to store a model pattern representing a plurality of predetermined features of a target object located in positions different from each other when the target object is viewed from a predetermined direction, and at least one partial region including a part of the model pattern;
• and
• a processor configured to
• extract the plurality of predetermined features from an image in which the target object is represented,
• calculate an entire coincidence degree representing a coincidence degree between the plurality of predetermined features set for the entire model pattern and the plurality of predetermined features extracted from a region corresponding to the model pattern on the image while changing a relative positional relationship between the image and the model pattern,
• calculate, for each at least one partial region, a partial coincidence degree representing a coincidence degree between one or more features included in the partial region among the plurality of predetermined features set for the entire model pattern and one or more of the predetermined features extracted from a region corresponding to the partial region on the image while changing the relative positional relationship, and
• determine that the target object is represented in a region on the image corresponding to the model pattern when the entire coincidence degree is equal to or greater than a predetermined entire coincidence degree threshold value and the partial coincidence degree calculated for each at least one partial region is equal to or greater than a predetermined partial coincidence degree threshold value,
• wherein
• a type of at least one feature of the plurality of predetermined features used for calculating the partial coincidence degree is different from a type of at least one feature of the plurality of predetermined features used for calculating the entire coincidence degree,
• the at least one partial region comprises a plurality of partial regions set in the model pattern, and
• the type of the at least one feature used for calculating the partial coincidence degree is different for each of the plurality of partial regions.