Image processing apparatus, imaging apparatus, driving assistance apparatus, mobile body, and image processing method

An imaging apparatus comprising:

• an imaging optical system;
• an image sensor configured to photoelectrically convert an object image acquired via the imaging optical system, and output an image signal;
• a first processor configured to perform image processing on the image signal;
• a second processor configured to perform correction processing to the image signal to which the image processing has been performed to correct optical distortion in the object image caused by the imaging optical system;
• a third processor configured to perform recognition processing on an image to which the image processing has been performed and in which the optical distortion has been corrected;
• and
• a fourth processor configured to superimpose a result of the recognition processing onto the image signal subjected to the image processing and the correction processing and output a resulting image signal,
• wherein the first processor is configured to perform at least two types of the image processing comprising:
• first image processing and second image processing,
• the second processor is configured to perform at least two types of the correction processing comprising:
• first correction processing and second correction processing,
• the third processor is configured to receive the image signal to which the first image processing and the first correction processing have been performed and output information indicating a location of a recognized object in the image signal,
• the fourth processor is configured to receive the image signal to which the second image processing and the second correction processing have been performed and the information indicating the location of the recognized object, and superimpose a drawing indicating the location of the recognized object onto a corresponding location in the image signal,
• the first image processing includes performing a near-linear contrast adjustment in order to minimize data loss during the recognition process,
• the second image processing includes increasing contrast or enhancing edges to thereby increase sharpness of the image,
• the first correction processing includes correcting the image by subdividing the image into a plurality of regions and performing distortion correction appropriate for the respective regions, and
• when the image is displayed in a super-wide angle, the second correction processing includes correcting height-direction distortion alone without correcting lateral-direction distortion, such that a viewer can see a three-dimensional object standing upright.

A driving assistance apparatus comprising:

• an imaging optical system;
• an image sensor configured to photoelectrically convert an object image acquired via the imaging optical system, and output an image signal;
• a first processor configured to perform image processing on the image signal;
• a second processor configured to perform correction processing to the image signal to which the image processing has been performed to correct optical distortion in the image caused by the imaging optical system;
• a third processor configured to perform recognition processing on the image to which the image processing has been performed and in which the optical distortion has been corrected;
• a fourth processor configured to superimpose a result of the recognition processing onto the image signal subjected to the image processing and the correction processing, and output a resulting image;
• and
• a display configured to display the resulting image output from the fourth processor,
• wherein the first processor can perform at least two types of the image processing comprising:
• first image processing and second image processing,
• the second processor is configured to perform at least two types of the correction processing comprising:
• first correction processing and second correction processing,
• the third processor is configured to receive the image signal to which the first image processing and the first correction processing have been performed and output information indicating a location of a recognized object in the image signal,
• the fourth processor is configured to receive the image signal to which the second image processing and the second correction processing have been performed and the information indicating the location of the recognized object, and superimpose a drawing indicating the location of the recognized object at a corresponding location in the image signal,
• the display is configured to display the resulting image upon receiving the resulting image from the fourth processor,
• the first image processing includes performing a near-linear contrast adjustment in order to minimize data loss during the recognition process,
• the second image processing includes increasing contrast or enhancing edges to thereby increase sharpness of the image,
• the first correction processing includes correcting the image by subdividing the image into a plurality of regions and performing distortion correction appropriate for the respective regions, and
• when the image is displayed in a super-wide angle, the second correction processing includes correcting height-direction distortion alone without correcting lateral-direction distortion, such that a viewer can see a three-dimensional object standing upright.

An image processing method comprising:

• acquiring an image;
• performing at least two types of image processing comprising:
• first image processing and second image processing, on the image;
• performing at least two types of correction processing comprising:
• first correction processing and second correction processing on the image to which the at least two types of image processing has been performed, to correct optical distortion in the image caused by an optical system that captured the image;
• performing recognition processing on the image to which the first image processing and the first correction processing have been performed;
• and
• superimposing a drawing indicating a location of a recognized object onto a corresponding location in the image subjected to the second image processing and the second correction processing, wherein
• the first image processing includes performing a near-linear contrast adjustment in order to minimize data loss during the recognition process,
• the second image processing includes increasing contrast or enhancing edges to thereby increase sharpness of the image,
• the first correction processing includes correcting the image by subdividing the image into a plurality of regions and performing distortion correction appropriate for the respective regions, and
• when the image is displayed in a super-wide angle, the second correction processing includes correcting height-direction distortion alone without correcting lateral-direction distortion, such that a viewer can see a three-dimensional object standing upright.