A gas detection device that gives a notification of a detected gas on the basis of a captured image obtained by capturing an image of a monitoring target, the gas detection device including: a gas detection unit that detects gas on the basis of the captured image and gives a notification of the detected gas; an input unit that receives input information from a user; a mask candidate region extraction unit that extracts a mask candidate region that is a candidate region of a mask region for which a notification of gas detection is suppressed; and a mask generation unit that generates mask data indicating the mask region, in which the gas detection unit gives a notification of a gas detected outside the mask region, and the mask generation unit generates, as the mask data, a region in which first mask candidate region information input from the input unit matches second mask candidate region information extracted by the mask candidate region extraction unit.

A connected component analysis (CCA) method, which can use labels repeatedly, comprising: defining a label pattern comprising a label and a plurality of neighboring labels; setting a center label of a current pixel of a target binary image according to a binary value of the current pixel and the neighboring pixels; setting at least two of the neighboring labels according to whether the current pixel is in any one of a first row, a first column and a last column; and recording the center label to a label buffer. Labels for marking pixels of the target binary image are first center labels, and then are second center labels, and are the first center labels again after the labels are the second center labels.

A method is disclosed for creating a precision mask for a utility pole and/or pole hardware (“utility pole”) depicted in a digital image. The utility pole is detected within a digital image comprising a plurality of pixels. An initial pixel mask is then created for the utility pole and determining a color range. The plurality of pixels is then processed within the digital image, wherein: the color of each pixel is sampled, and each pixel is included or excluded from the precision mask based at least partially on its sampled color and the color range; and outputting the precision mask. Other disclosed methods provide for identification of a class of utility pole depicted in a digital image by comparison to a plurality of 3D models. Still other disclosed methods may be used to predict the height of a utility pole depicted in a digital image.

The present invention relates to a grabbing detection method based on an RP-ResNet, which method belongs to the field of computer vision, and in particular relates to recognition and positioning of a grabbing point of a mechanical arm. The method comprises: inputting a target object image; pre-processing data; performing data processing by means of an RP-ResNet model; and finally, generating a grabbing block diagram of a grabbing target. On the basis of a model ResNet 50, a region proposal network is used in the 30th layer of a network, fuzzy positioning is performed on the position of a grabbing point, feature information of high and low layers is fully fused to strengthen the utilization of information of low layers, and an SENet structure is added to the 40th layer of the network, thereby further increasing the detection accuracy of a grabbing point. By means of a grabbing detection framework based on ResNet-50, a residual network, a region proposal idea and SENet are combined, such that it is ensured that rapid target detection is realized, and the accuracy rate of target detection is further improved.

A license plate identification method is provided, including the following steps of: obtaining a to-be-processed image; obtaining a plurality of feature maps including target features through a feature map extraction module; obtaining at least one region including the target feature in each feature map and giving each frame of each feature map scores corresponding to the target features through a target location extraction module; classifying each frame in each feature map according to the scores through a target candidate classification module and retaining at least one region that corresponds to character features; and obtaining a license plate identification result according to the region that corresponds to the character feature through a voting/statistics module.

Methods and apparatuses for detecting one or more objects (e.g., dropped objects) by a robotic device are described. The method comprises receiving a distance-based point cloud including a plurality of points in three dimensions, filtering the distance-based point cloud to remove points from the plurality of points based on at least one known surface in an environment of the robotic device to produce a filtered distance-based point cloud, clustering points in the filtered distance-based point cloud to produce a set of point clusters, and detecting one or more objects based, at least in part, on the set of point clusters.

A medical image processing apparatus of an embodiment includes processing circuitry. The processing circuitry receives a medical image of a target region. The processing circuitry generates an image pair including a local image having local features of the target region and a global image having global features of the target region on the basis of the received medical image. The processing circuitry performs segmentation and classification of the target region on the image pair by a neural network.

Provided is an apparatus for recognizing an object that includes an object inference module configured to process an original image captured by a camera module and generate an image of a size to be input to a machine learning inference model, wherein the object inference module includes the machine learning inference model, and outputs a result of recognition and classification of an object inferred through the machine learning inference model, and the machine learning inference model processes an input image to infer an object included in the input image.

In order to perform quantitative analysis on an object in an image, it is important to accurately identify the object, but when plural objects are in contact with each other, it is potential that a target portion cannot be accurately identified. An image is segmented into a foreground region and a background region, the foreground region being a region in which an object for which quantitative information is to be calculated is shown, and the background region being a region other than the foreground region. With respect to a first object and a second object in contact with each other in the image, a contact point between the first object and the second object is detected based on a region segmentation result output by a segmentation unit. The first object and the second object can be separated by connecting two boundary reference pixels including a first boundary reference pixel that is a pixel in a background region closest to the contact point, and a second boundary reference pixel that is a pixel in a background region in a direction opposite to the first boundary reference pixel across the contact point.

The invention relates to a method for determining the particle size distribution of parts of a bulk material (2) fed onto a conveyor belt (1), wherein a depth image (6) of parts of the bulk material (2) is captured in a capturing region (4) by means of a depth sensor (3). In order to reliably classify bulk material at conveying speeds of more than 2 m/s even if there are overlaps, without having to take structurally complicated measures for this purpose, according to the invention, the captured two-dimensional depth image (6) is fed to a convolutional neural network, which has been trained in advance and which has at least three convolutional layers lying one behind the other and one downstream amount classifier (22) per class of a particle size distribution, the output values (21) of which amount classifiers are output as the particle size distribution of the bulk material present in the capturing region (4).