The present invention provides an apparatus and method for image classification and segmentation based on a feature-guided network, a device, and a medium, and belongs to the technical field of deep learning. A feature-guided classification network and feature-guided segmentation network of the present invention include basic unit blocks. A local feature is enhanced and a global feature is extracted among the basic unit blocks. This resolves a problem that features are not fully utilized in existing image classification and image segmentation network models. In this way, a trained feature-guided classification network and feature-guided segmentation network have better effects and are more robust. The present invention selects the feature-guided classification network or the feature-guided segmentation network based on a requirement of an input image and outputs a corresponding category or segmented image, to resolve a problem that the existing classification or segmentation network model has an unsatisfactory classification or segmentation effect.

An image processing method includes segmenting, by a cloud server and based on a zoom ratio of a high-definition image in a high-definition image library, a low-definition image of a small zoom ratio from a terminal to obtain a plurality of image blocks. For each image block, performing, by the cloud server, retrieval and matching in the high-definition image library to obtain one or more high-definition reference images, and then performing, by the cloud server, image quality improvement using an image enhancement network that is based on the high-definition reference images to obtain a plurality of high-definition image blocks. Finally, stitching, by the cloud server, the high-definition image blocks into a super-definition image with higher resolution and higher definition, and returning, by the cloud server, the super-definition image to the terminal.

The present invention relates to a method of extracting a roof edge image for solar panel installation by using machine learning, the method comprising: a training step for passing original rooftop image data through a second generation unit of an image extraction system to output an image similar to a target image, and passing image data, from which a rooftop edge has been detected, through a first generation unit of the system to identify the image data from an original image; a step for segmenting an obstruction hiding a roof edge, and receiving, by a second discriminator unit, an image in which the roof edge has been detected; a step for optimizing the weight of a parameter, and training the second generation unit and the second discriminator unit again; and a step for automatically connecting edge portions after extracting edges, and generating a complete roof edge image.

Provided are an image exposure adjustment method and apparatus, a device, and a medium. The image exposure adjustment method includes performing human body detection on a collected image; in a case where a human body is detected, segmenting the image to determine a foreground region and a background region in the image; determining a mask image according to the foreground region and the background region; and determining an exposure weight table according to the mask image, and performing exposure value adjustment on the image according to the exposure weight table.

Provided is an object recognition device capable of accurately estimating a distance from an own vehicle to an object such as a pedestrian. There are provided: a part specification unit 104 that determines a specific part region in an object by analyzing an image of a region corresponding to the object detected by an overlapping region object detection unit 103; an image information storage unit 105 that stores image information on the part region determined by the part specification unit 104; a three-dimensional information acquisition unit 106 that acquires three-dimensional information of the part region specified by the part specification unit 104 on the basis of the distance information acquired by a distance information acquisition unit 102; a non-overlapping region object detection unit 107 that detects the part region specified by the part specification unit 104 with reference to the image information stored in the image information storage unit 105 for the non-overlapping region; and a distance calculation unit 108 that calculates a distance to an object including the part region on the basis of the detection region information on an image of the part region detected by the non-overlapping region object detection unit 107 and the three-dimensional information acquired by the three-dimensional information acquisition unit 106.

A method for automatically recognizing a location of an object so that a robot can grasp and transfer the product in an apparatus for automatically recognizing a location of an object is provided. The method for automatically recognizing a location of an object obtains one frame from image data of the product produced by the manufacturing facility, detects location and shape information of the product based on a hue, saturation, value (HSV) color space using the frame, generates a recipe for the robot to grasp and transfer the product based on the location and shape information of the product, and transmits the recipe to the robot.

A personal authentication apparatus comprises a first image capture unit that captures an eyeball image, a first image processing unit that obtains brightness information of the captured eyeball image and generates divided eyeball images by dividing a region of the eyeball image based on the brightness information of the eyeball image, a first storage unit that stores brightness information of the divided eyeball images and position information of the divided eyeball images in relation to the eyeball image in association with each other, and a learning unit that executes different types of learning processing respectively for the divided eyeball images in accordance with the brightness information and the position information of the divided eyeball images.

Provided are an image fusion classification method and device. The method includes that: a three-dimensional weight matrix of a hyperspectral image is obtained by use of a Support Vector Machine (SVM) classifier (101); superpixel segmentation is performed on the hyperspectral image to obtain K superpixel images, K being a positive integer (102); the three-dimensional weight matrix is regularized by use of a superpixel-image-based segmentation method to obtain a regular matrix (103); and a class that a sample belongs to is determined according to the regular matrix (104).

Device (100) for attesting the operations of an in-vitro diagnostic device (50) comprising: a block (101) for capturing a plurality of frames of the tip (51); a block (102) for storing the plurality of frames; a block (103) for evaluating the right hooking of the tip (51) to the in-vitro diagnostic device (50); a block (104) for evaluating the volume of a liquid contained in the tip (51); a block (105) for carrying out verification before the operation of dispensing the liquid; a block (106) for carrying out a post-dispensing verification; blocks (107, 108) for emitting electronic signals; a block (109) for integrating a system for managing the errors; a block (110) for saving the data; a block (111) for communicating with the in-vitro diagnostic device (50).

It is an object of the present invention to provide an object collating device and an object collating method that enable matching of images of a dividable medical article with desirable accuracy and easy confirmation of matching results. In the object collating device according to the first aspect, when the object is determined to be divided, the first image for matching is collated with the image for matching (the second matching image) for the objects in the undivided state, so that the region to be matched is not narrowed, and matching of the images of the dividable medical article is achieved with desirable accuracy. In addition, since the first and second display processing is performed on the images for display determined to contain the objects of the same type, matching results can easily be confirmed.