Image processing apparatus and image processing method
Provided are an image processing apparatus and an image processing method for improving the accuracy of a recognition result of a current object included in captured images. The image processing apparatus performs recognition processing on the current object on the basis of recognition results of the current object obtained from a plurality of captured images with different output information regarding imaging so that the accuracy of the recognition result of the captured images can be improved. For example, the image processing method can be applied to an electronic device having a function of performing the recognition processing on the current object.
1. An image processing apparatus, comprising:
a central processing unit (CPU) configured to:
acquire a first plurality of captured images of a specific object from a first imaging element, wherein
the first plurality of captured images is a plurality of image frames generated in time series,
first output information of the first plurality of captured images is different from second output information of a second plurality of captured images of the specific object,
the second plurality of captured images of the specific object is captured by a second imaging element,
the first output information is associated with an imaging operation of the first plurality of captured images, and
the second output information is associated with an imaging operation of the second plurality of captured images;
execute a recognition process on the first plurality of captured images of the specific object;
output a first recognition result of the specific object based on the execution of the recognition process; and
synthesize the first recognition result and a second recognition result of the specific object, wherein the second recognition result of the specific object is based on the second plurality of captured images.
2. The image processing apparatus according to claim 1 , wherein
the first output information indicates at least one of a resolution of each captured image of the first plurality of captured images or a frame rate of each captured image of the first plurality of captured images.
3. The image processing apparatus according to claim 2 , wherein the CPU is further configured to execute the recognition process on the first plurality of captured images based on the second recognition result of the specific object in the second plurality of captured images.
4. The image processing apparatus according to claim 1 , wherein the CPU is further configured to:
determine, based on threshold information, that the first recognition result and the second recognition result are to be synthesized; and
synthesize the first recognition result and the second recognition result based on the determination.
5. The image processing apparatus according to claim 4 , wherein the threshold information indicates one of:
an allowable range of deviation between a position of the specific object in the first recognition result and a position of the specific object in the second recognition result, or
an allowable range of a difference between a type of the specific object in the first recognition result and a type of the specific object in the second recognition result.
6. The image processing apparatus according to claim 1 , wherein the CPU is further configured to:
adjust at least one of the first recognition result or the second recognition result such that the at least one of the first recognition result or the second recognition result is in a synthesizable state; and
synthesize the first recognition result and the second recognition result after the adjustment.
7. The image processing apparatus according to claim 1 , further comprising a memory configured to store the first recognition result,
wherein the CPU is further configured to synthesize the stored first recognition result and the second recognition result at a specific time.
8. The image processing apparatus according to claim 1 , wherein the CPU is further configured to:
set delay information for a time required for execution of the recognition process on the second plurality of captured images; and
execute the recognition process on the first plurality of captured images based on the set delay information.
9. An image processing method, comprising:
acquiring a first plurality of captured images of a specific object from a first imaging element, wherein
the first plurality of captured images is a plurality of image frames generated in time series,
first output information of the first plurality of captured images is different from second output information of a second plurality of captured images of the specific object,
the second plurality of captured images of the specific object is captured by a second imaging element,
the first output information is associated with an imaging operation of the first plurality of captured images, and
the second output information is associated with an imaging operation of the second plurality of captured images;
executing a recognition process on the first plurality of captured images of the specific object;
outputting a first recognition result of the specific object based on the execution of the recognition process; and
synthesizing the first recognition result and a second recognition result of the specific object, wherein the second recognition result of the specific object is based on the second plurality of captured images.
10. A non-transitory computer-readable medium, having stored thereon computer-executable instructions, which when executed by a processor, cause the processor to execute operations, the operations comprising:
acquiring a first plurality of captured images of a specific object from a first imaging element, wherein
the first plurality of captured images is a plurality of image frames generated in time series,
first output information of the first plurality of captured images is different from second output information of a second plurality of captured images of the specific object,
the second plurality of captured images of the specific object is captured by a second imaging element,
the first output information is associated with an imaging operation of the first plurality of captured images, and
the second output information is associated with an imaging operation of the second plurality of captured images;
executing a recognition process on the first plurality of captured images of the specific object;
outputting a first recognition result of the specific object based on the execution of the recognition process; and
synthesizing the first recognition result and a second recognition result of the specific object, wherein the second recognition result of the specific object is based on the second plurality of captured images.