9results about How to "Improve visual recognition" patented technology

The invention relates to the technical field of single cell data processing, in particular to a database visualization method based on single cells. The method comprises the following steps: acquiring a single cell data set; standardizing the single cell data set to obtain standardized single cell data; performing high-dimensional feature extraction on the standardized single-cell data to obtain single-cell high-dimensional feature data; and performing nonlinear dimension reduction processing on the single-cell high-dimensional feature data to obtain single-cell dimension reduction data. Performing data clustering on the single cell dimension reduction data to obtain single cell clustering data; performing gene differential expression extraction on the single cell clustering data to obtain differential expression gene data; multi-modal data fusion analysis is carried out on single cell data through a data processing technology, a feature mapping technology and a visualization technology; and dynamic weight division is carried out on the single cell database, so that the effect of visualization of the single cell database is enhanced.

The invention discloses an industrial robot 3D visual identification method based on deep learning. The method comprises the following steps: step 1, carrying out data acquisition and preprocessing; step 2, extracting data features; 3, selecting and constructing a deep learning model; 4, training and optimizing the model; 5, evaluating and selecting the model; 6, deploying and applying the model; 7, continuously improving and maintaining the model; the industrial robot 3D visual identification method based on deep learning is an advanced industrial automation technology, the advantages of a deep learning algorithm and the advantages of a 3D visual technology are combined, the complete process of the industrial robot 3D visual identification method based on deep learning is formed through the steps, and the industrial robot 3D visual identification method based on deep learning is realized from data acquisition to model deployment and application. And through continuous improvement and maintenance, higher and faster identification precision and adaptability are provided for the industrial robot, and the technology plays a more and more important role in development of industrial automation and intelligence in the future.

Provided is a selector switch device for a vending machine, which improves the visibility of a display pattern when a translucent member receives direct sunlight. A selection switch device (20) is provided with: a light-emitting element; a display sheet (33) for displaying information relating to the corresponding product by receiving the light emitted from the light-emitting element in a display pattern; and a selection button (23) provided on the front surface side of the display sheet (33), the display pattern of the display sheet (33) being formed by forming a light-transmissive white layer (33c) having a color with high light transmissivity on the back surface side or the front surface side of the transparent substrate (3a).

The application discloses a kind of based on big data visualization's transformer substation intelligent monitoring method, it is related to measurement data visualization technical field, including, the offset degree between the space-time fingerprint base template and real-time behavior fingerprint is calculated, generates offset vector set, obtains the distribution of offset vector set in space by kernel density algorithm, generates abnormal density field and space abnormal density point cloud;According to space abnormal density point cloud setting density threshold and solving the equipotential surface of abnormal density field, generate abnormal three-dimensional equipotential surface group;Extract risk evaluation index of abnormal three-dimensional equipotential surface group and carry out dynamic rendering to abnormal three-dimensional equipotential surface group, output dynamic visualization picture.The application constructs space-time fingerprint base template and calculates offset vector set, and abstract offset behavior is converted into three-dimensional risk form with space boundary, expansion trend and density aggregation characteristics, improves the explainability of transformer substation abnormal structure, visual identification and forward-looking monitoring capability.

The municipal engineering construction fence with the expansion structure comprises a fence body, a mounting frame is connected to the interior of the top end of the fence body in a sliding mode, a protective net is mounted in the mounting frame, adjusting mechanisms are symmetrically arranged in the fence body, and a light reflecting assembly is arranged on one side of the fence body; the adjusting mechanism comprises mounting boxes symmetrically and fixedly connected to the interior of the fence body, mounting pipes are fixedly connected to the interiors of one ends of the mounting boxes, pull rods are slidably connected to the interiors of the mounting pipes, pull plates are fixedly connected to one ends of the pull rods, and inserting blocks are fixedly connected to one sides of the pull plates. And fixing pipes are symmetrically and fixedly connected to the interior of one end of the mounting box, so that the effect of accurately adjusting the height of the protective net can be achieved, the effect that the fence body can be expanded is achieved, the fence can flexibly adapt to various construction environments and can be accurately adjusted to cover construction areas with different heights, and the construction efficiency is improved.

The invention provides a high-frequency module and a communication device. A high-frequency module (1) is provided with: a module substrate (91) having a main surface (91a); first and second circuit components disposed on the main surface (91a); a resin member (92) covering the main surface (91a) and at least a portion of the first and second circuit components; a metal shield layer (95) that covers at least the upper surface (92a) of the resin member (92); and a metal shield plate (70) disposed on the main surface (91a) and disposed between the first circuit member and the second circuit member in a plan view of the main surface (91a). And the metal shielding plate (70) is connected with the metal shielding layer (95). The upper surface (92a) of the resin member (92) is provided with an imprint part (80) which shows given information. The engraved portion (80) does not overlap with at least the upper end surface (70a) of the metal shield plate (70) in a plan view of the main surface (91a).

The invention provides a tea leaf composition, pink tea soup and a preparation method, and belongs to the technical field of food processing. The invention provides a tea composition. The tea composition comprises rosemary and Biluochun. The mass ratio of the rosemary to the Biluochun is (2: 3)-(1: 1). The pink tea soup can be obtained by brewing the tea combination, has unique tea soup color and luster, and is beneficial to enhancing the visual recognition degree of brands. Moreover, the rosemary and Biluochun in the tea composition are natural edible plants, so that pink tea soup obtained by brewing the tea composition has the advantages of no additive, natural components, health and safety, and can meet the requirements of consumers at the current stage on healthy diet.

A screen saver controller, a display device, and a method for operating the display device are provided. The screen saver controller includes: a load calculator for generating load data for each of a plurality of panel blocks included in a display panel based on input image data; a lifespan calculator for generating lifespan data for each of the plurality of panel blocks based on accumulated degradation stress values ​​in the display panel; a first logic circuit for receiving the load data and the lifespan data, and generating operation data based on the load data and the lifespan data; and a second logic circuit for receiving the input image data and the operation data, and generating screen saver data based on the input image data and the operation data. When operating in a screen saver mode for performing screen saver operations, the screen saver controller adjusts the brightness of the display panel based on the screen saver data.