A store monitoring system includes: an imaging apparatus; and a store monitoring apparatus that monitors a display shelf in a store by using a captured image captured by the imaging apparatus, the store monitoring apparatus including: an acquisition unit that sequentially obtains the captured image from the imaging apparatus; and an image generation unit that extracts from the captured image a shelf image in which the display shelf appears, by eliminating from the captured image a non-shelf image in which the display shelf does not appear, and that generates a front shelf image obtained when the display shelf is imaged from a front, by correcting a distortion of the shelf image.

System and methods are disclosed for capturing license plate (LP) information of a vehicle in relative motion to a camera device. In one example, the camera system detects the LP in multiple frames, then aligns and geometrically rectifies the image of the LP by scaling, warping, rotating, and/or performing other functions on the images. The camera system may optimize capturing of the LP information by executing a temporal noise filter on the aligned, geometrically rectified images to generate a composite image of the LP for optical character recognition. In some examples, the camera device may include an image sensor, such as a high dynamic range (HDR) sensor, modified to set long and short exposures of the HDR sensor to capture frames of a vehicle's LP, but without consolidating the images into a composite image. The camera system may set optimal exposure settings based on detected relative speed of the vehicle.

System and methods are disclosed for capturing license plate (LP) information of a vehicle in relative motion to a camera device. In one example, the camera system detects the LP in multiple frames, then aligns and geometrically rectifies the image of the LP by scaling, warping, rotating, and/or performing other functions on the images. The camera system may optimize capturing of the LP information by executing a temporal noise filter on the aligned, geometrically rectified images to generate a composite image of the LP for optical character recognition. In some examples, the camera device may include an image sensor, such as a high dynamic range (HDR) sensor, modified to set long and short exposures of the HDR sensor to capture frames of a vehicle's LP, but without consolidating the images into a composite image. The camera system may set optimal exposure settings based on detected relative speed of the vehicle.

Provided are a method and an apparatus for restoring an object image, capable of restoring an image naturally by detecting positions of landmarks of an object in a bounding-box detected from an input image, performing warping to align the object at a central position or a reference position on the basis of the landmarks, improving the image using a learning model learned from the aligned object image, performing inverse warping for rotating the improved object image in an original direction or at an original angle, and inserting the inversely-warped object image into the input image. In addition, provided are a method and an apparatus for restoring an object image, capable of detecting positions of landmarks of an object in a bounding-box detected from an input image, performing pose estimation for a side object on the basis of the landmarks, and improving an image using a learning model learned from a side object image corresponding to the pose estimation result.

Provided are a method and a device for restoring a facial image, capable of restoring an image naturally by detecting positions of landmarks of a face in a bounding-box detected from an input image, improving an image using a learning model learned from a front facial image after performing warping for aligning a front face to be positioned at a central position or a reference position on the basis of the landmarks, performing inverse warping for rotating the improved image in an original direction or at an original angle, and inserting the inversely-warped image into the input image. In addition, provided are a method and a device for restoring a facial image, capable of performing pose estimation for a face in a bounding-box detected from an input image, and improving the image using a learning model learned from a side facial image corresponding to a result of the pose estimation.

Disclosed are a method and device for determining an icon position. The method includes: detecting a target object in a target image and determining the reference position of the target object in the target image, and detecting a salient position in the target image, thereby obtaining the reference position of a key target or object in the target image, and a salient position possibly requiring more attention in the target image; and selecting, according to the distance between the reference position or salient position and preset candidate positions, an icon position from the candidate positions.

Disclosed are a method and device for determining an icon position. The method includes: detecting a target object in a target image and determining the reference position of the target object in the target image, and detecting a salient position in the target image, thereby obtaining the reference position of a key target or object in the target image, and a salient position possibly requiring more attention in the target image; and selecting, according to the distance between the reference position or salient position and preset candidate positions, an icon position from the candidate positions.

There is provided an image processing device capable of easily reducing a burden related to image recognition and obtaining stable recognition performance even when a posture of a vehicle changes from moment to moment. An image processing device 50 includes a storage unit 5 and an image processing unit 6. The storage unit 5 stores a pickup image acquired by an image pickup unit 4 mounted on the vehicle and an inclination angle transmitted from an inclination angle sensor 2 that detects the inclination angle of the vehicle. The image processing unit 6 performs image processing on the pickup image. The image processing unit 6 associates the pickup image in an inclined state with the inclination angle based on a time difference Δt between a time from when the pickup image in the inclined state is acquired by the image pickup unit 4 until the acquired pickup image is stored in the storage unit 5, and a time from when the inclination angle is detected by the inclination angle sensor 2 until the detected inclination angle is stored in the storage unit 5. The image processing unit 6 generates a recognition image in a non-inclined state, which is used for image recognition of the pickup image, from the pickup image in the inclined state based on the inclination angle associated with the pickup image.

According to the present disclosure, a handwriting image and a background image are combined, thereby generating a combined image, a correct answer label image is generated based on the handwriting image, and the generated combined image and the generated correct answer label image are used as learning data for training a neural network.

According to the present disclosure, a handwriting image and a background image are combined, thereby generating a combined image, a correct answer label image is generated based on the handwriting image, and the generated combined image and the generated correct answer label image are used as learning data for training a neural network.