Abstract: The present invention discloses a method and an apparatus for extracting mountain landscape buildings based on high-resolution remote sensing images. The method comprises: segmenting a remote sensing image, and extracting non-vegetation areas from the remote sensing image by using NDVI; segmenting the non-vegetation areas, and extracting building areas by using NDBI; segmenting the building areas again, and calculating a normalized difference build shadow index NSBI of each patch; calculating NSBI separator of each patch in the non-vegetation areas and setting a separator threshold, and extracting landscape building areas based on the threshold. In the present invention, by introducing a near infrared band in the remote sensing image spectrum, in which there is a significant difference between shadows and non-shadows, the influence of large shadow areas in mountainous shady areas in the remote sensing image on the result of extraction is reduced.

Abstract: A spray gun shower head includes a shower body, a switching body and a reset assembly. The shower body has a water intake path and at least one first and second water outflow paths. The switching body is movable between first and second switch positions and configured to switch the water intake path to be communicated with the first or second water outflow path. The reset assembly includes at least one reset member and a push block. The second water outflow path is hidden in the shower body through a cover. The cover is pushed against by an opening structure to be opened. The reset member is disposed on the shower body, and one end thereof cooperates with the push block. The push block is disposed on the switching body and rotates toward the second switch position along with the switching body to press against the reset member.

Abstract: Provided are a training method of a deep convolutional neural network, an abnormal behavior monitoring method and system, a storage medium and computer device. The deep convolutional neural network is a single-stage dual-branch convolutional neural network, and includes a first branch for predicting confidences and a second branch for predicting part affinity vector fields. The method includes: inputting an image to be identified; according to one or more preset objects to be identified, performing feature analysis on the image to be identified to obtain one or more feature map sets for the one or more objects to be identified in the image to be identified, wherein each feature map set corresponds to one object to be identified; inputting a feature map set into the first branch to obtain confidence prediction results; inputting the confidence prediction results and feature map set into the second branch to obtain affinity field prediction results.

Abstract: A shower device lifting rod includes a first rod member including a first connection portion disposed at one end thereof, a second rod member including a second connection portion disposed at one end thereof, and a connection member including a third connection portion and a fourth connection portion disposed at two ends thereof. The third connection portion is fixedly connected to the first connection portion after being inserted in the first connection portion, and the fourth connection portion is fixedly connected to the second connection portion after being inserted in the second connection portion. The members may be separated when the shower device lifting rod adopting such structure is packaged and transported and may then be put together when the shower device lifting rod is actually assembled. In this way, a package volume is reduced and transportation costs are lowered.

Abstract: Provided is a hair cutting assembly. The hair cutting assembly includes at least one first shearing wheel, first transmission shaft that drives the first shearing wheel to rotate, at least one second shearing wheel and second transmission shaft that drives the second shearing wheel to rotate. Opposite end surfaces of each first shearing wheel and the corresponding second shearing wheel are attached to each other. The first cutting blades, formed by intersecting end surface of the first shearing wheel with side wall, and the second cutting blades, formed by intersecting end surface of the second shearing wheel with side wall, oppositely move in a feed direction, such that long and thin linear sundries such as hairs and fabrics can be continuously shorn. A floor drain with the hair cutting assembly is also provided.

Abstract: The disclosure discloses a shower head holder structure, which includes a shower head and a holder, the shower head has a lever and a pair of blocks; when the first end of the lever is pressed, the second end of the lever on the back of the nozzle is lifted, and the two blocks on both sides of the second end are brought closer to each other; when the first end of the lever is released, the two blocks are away from each other; the holder is provided with a pair of slots matched with the pair of blocks; when the blocks are inserted into the pair of slots, the shower head is hung up on the holder.

Abstract: The disclosure discloses an out-of-wall water discharging device, including connecting pipes, a fixing base, a rotating base rotatably connected to a fixing base through the clutch assembly, a rotating tube rotatably connected to the rotating base through the connecting element, and a water discharging member rotatably connected to the rotating tube through a ball head element communicating with each other. The above solution can achieve rotation in at least three axial directions to adjust the angle of the water discharging end.

Abstract: Techniques for evaluating performance of reviewing events in a network community are described herein. The disclosed techniques include authorizing a first user account to review the events in the network community for a first valid period; monitoring information associated with the first user account reviewing the events in the network community; generating data indicative of review behaviors associated with the first user account during at least a part of the first valid period based at least in part on the information and predetermined parameters of evaluating performance; determining values of a plurality of evaluation coefficients based on the generated data and predetermined rules; and determining a performance evaluation result for the first user account based on the generated data and the values of the plurality of evaluation coefficients.

Abstract: A urinal and a urinal flushing control method. The urinal includes a urinal body, an electromagnetic valve for turning on and off flushing, a capacitive sensor for sensing a liquid amount on an outer wall of the urinal body, and a controller for receiving a capacitance value signal to control on and off of the electromagnetic valve. The controller sets a flushing timing reference point when determining that a capacitance value exhibits a dropping trend after a first period since the electromagnetic valve is controlled to be turned off, and controls to turn on the electromagnetic valve when the flushing timing reference point is present for a second period. The above solution can reduce the chance of erroneous flushing and facilitate the adaptability respect to the glazed surface of the urinal body.

Abstract: Techniques for managing and presenting violation events in a network community are described herein. The disclosed techniques include receiving a file containing information indicative of a violation by a user account in the network community who violates community guidelines and storing the file into a violation display pool in response to a determination that a corresponding violation score is greater than a predetermined threshold. The violation display pool comprises a plurality of violation events that each comprise information indicative of a source of a corresponding violation event, at least one attribute of content associated with the corresponding violation event, and information relating to content of the corresponding violation event. A storage server sends the plurality of violation events to a plurality of client computing devices for display via an interface.

Abstract: A hybrid shower head includes: a first shower head including a first shower head portion and a positioning portion; a second shower head including a second shower head portion and a matching portion which is disengaged from or engaged with the positioning portion; a two-position three-way diverter valve including a first water outlet communicating with the first shower head portion and a second water outlet communicating with the second shower head portion; and a switch controlling the two-position three-way diverter valve to select a water path. When the positioning portion is engaged with the matching portion, water is automatically discharged from only one of the first water outlet and the second water outlet. When the positioning portion is disengaged from the matching portion, water is automatically discharged from the other of the first water outlet and the second water outlet.

Abstract: Techniques for identifying rule violations in a network community are described herein. The disclosed techniques include receiving a report event from a first client computing device, wherein the report event is associated with content and a possible violation and the content comprises one or more comments relative to a plurality of video frames; identifying the content based on the possible violation; sending the content to a plurality of review users, wherein the content is presented and viewable via an interface by each review user; determining whether a voting instruction has been received from each review user during a predetermined period of time; recording a voting instruction from a first review user in a list of voting instructions in response to a determination that the voting instruction has been received during the predetermined period of time; and identifying the possible violation as a violation based on the list of voting instructions.

Abstract: A water pre-storable toilet includes a drain valve located between an injection duct and a water tank, and a float connected to the drain valve is further provided in the water tank. The float is configured to apply an upward pulling force to the drain valve at least when a water level of the water tank reaches a maximum. The float is also configured in such a way that when the float applies the upward pulling force to the drain valve, a buoyancy of the float may not overcome a water pressure caused by the water in the water tank to the drain valve, a weight of the drain valve itself, and a negative pressure formed when the pre-stored water in the injection duct reaches the highest water level regardless of water level of a water in the water tank and thus can't open the drain valve.

Abstract: A water separator with a shower seat includes a valve body, a frictional unit, a shower seat, and a water outflow switching mechanism. The valve body is provided with a plurality of snap-fit joints disposed circumferentially. The frictional unit is mounted on the valve body and an outer wall thereof is provided with a first frictional surface. The shower seat is provided with a connecting chamber having an open end, and a snap-fit surface and a second frictional surface are circumferentially disposed in the connecting chamber, respectively. The water outflow switching mechanism is mounted in the valve body. Hook heads of the snap-fit joints are snapped to the snap-fit surface to prevent the shower seat from getting separated from the valve body. When the shower seat rotates with respect to the valve body, the first frictional surface and the second frictional surface rotate accordingly and interact with each other.

Abstract: Provided is a hanging connection sliding base, including a sliding base body sleeved over a rod body, a first elastic element, a packing block and a lever. The lever is hinged to the sliding base body, and is pulled to drive the packing block to be away from the rod body. The first elastic element is arranged between the packing block and the sliding base body, and the packing block is driven to tightly urge the rod body when the lever is released. A related hanging connection frame is further provided. The hanging connection frame includes the above hanging connection sliding base and a rod body. Two sides of the rod body are provided with sliding grooves for receiving the hanging connection sliding base. By the hanging connection sliding base, a ground clearance of the hanging connection sliding base can be adjusted with a single hand.

Abstract: A fall detection method, device, and system are disclosed in this application for detecting whether a target object falls in a detection area. The fall detection method includes: receiving a WIFI signal transmitted by a transmitter in the detection area and extracting CSI data from the WIFI signal; preprocessing the CSI data to obtain CSI data to be identified, and processing the CSI data to be identified through a deep neural network to determine whether the target object falls in the detection area. In this application, a deep neural network is adopted to perform fall detection and the detection accuracy is improved.

Abstract: A detection method, a detection device, a terminal, and a detection system are provided, for detecting a state of a target object in a detection area. The detection method includes: filtering a millimeter-wave radar signal received in the detection area; and extracting features suitable for indicating a motion mode of the target object in the detection area from each frame of the filtered millimeter-wave radar signal; monitoring an initial change point of the features through a flow window; caching a predetermined number of features starting from the initial change point; and identifying the cached features by a classifier to determine the state of the target object in the detection area.