32results about How to "Reduce the chance of confusion" patented technology

The invention relates to an intelligent sorting garbage bin system and an intelligent sorting garbage bin based on the Internet of Things, belonging to the technical field of the Internet of Things. The intelligent sorting garbage bin system includes multiple intelligent sorting garbage bins and a remote server; the smart sorting garbage bin includes a control unit and a communication unit; when garbage is detected, the image of the current garbage is obtained; the current garbage type is identified based on the local garbage recognition model , if it cannot be identified, it will be sent to the remote server to use the remote garbage identification model for identification; if the remote server cannot be identified, the pickup will be turned on to obtain the garbage name information; the garbage type will be identified based on the garbage name information; based on the type identification result, control the garbage classification The picking actuator guides the current garbage to the corresponding sub-garbage storage box. The intelligent sorting garbage bin system is based on the communication technology of the Internet of Things to use the remote server to help identify, thereby effectively improving the accuracy of garbage classification, and can be widely used in garbage disposal and other fields.

The invention provides an insulated framework, a motor stator and a motor. The insulated framework comprises an annular portion (101) and a plurality of tooth portions (102) projecting from the inner wall of the annular portion (101), wherein opening grooves are formed in the annular portion (101) on both sides of a joint position of at least one of the plurality of tooth portions (102) and the annular portion (101) in the circumferential direction; opening grooves are further formed on one side of joint positions of the rest of tooth portions (102) and the annular portion (101) in the circumferential direction; and the opening grooves on both sides in the circumferential direction of the at least one tooth portion (102) having the opening grooves on both sides in the circumferential direction are of the same structure. By adopting the insulated framework, the motor stator and the motor, one of the two opening grooves is used when in need of serial wire winding, and the other is used when in need of parallel wire winding, so that the single insulated framework can achieve the functions and effects of serial wire winding and parallel wire winding of the stator simultaneously, and the applicability is effectively improved.

The embodiment of this specification discloses a keep-alive system for http long connections. The management and control server controls each keep-alive thread to perform heartbeat detection according to the corresponding target heartbeat interval, that is, it can be determined that the management and control server controls the thread corresponding to the heartbeat detection, and The management and control server also controls the threads in the first thread pool in the business server, so that the management and control server can uniformly control the threads in the business server. When the threads are under unified control, it can effectively reduce the number of business requests and heartbeat requests corresponding to the same http The probability of connection occurrence can reduce the probability of data packet confusion and increase the probability of normal operation of leased, available and expired http connections.

A COB manufacturing method, including a substrate manufacturing step, a solid crystal step, and a wire bonding step, the wire bonding step: (1) among several LED chips arranged in a line, the first wire arc bonding is used between adjacent LED chips connected to form the first series branch; (2) the head and tail between the adjacent first series branches are connected by the second arc welding wire to form the second series branch; (3) in the two rows of LED chips arranged in a staggered manner, the phase Adjacent LED chips are connected by the second wire arc welding wire to form the third series branch; (4) The ends between the second series branch and the third series branch are connected by the second wire arc welding wire to form the fourth series connection (5) The positive and negative poles of the fourth series branch and the pad are connected by the third wire arc welding wire. Since the three wire arc welding wires are welded separately and independently, the frequency of changing the welding wire process of the wire welding machine is reduced, thereby improving the welding efficiency and reducing the probability of confusion.

The invention discloses a method and equipment for extracting mountain landscape buildings based on high-resolution remote sensing images. The method includes: segmenting the remote sensing image, using NDVI to extract the non-vegetation area in the remote sensing image; segmenting the non-vegetation area, extracting the building area through NDBI; segmenting the building area again, and calculating the normalized shadow of each patch Building index NSBI; calculate the NSBI discrimination of each patch in the non-vegetated area and set the discrimination threshold, based on which the landscape architecture area is extracted. The present invention reduces the influence of the large-area shadow area in the shaded area of ​​the remote sensing image on the extraction result by introducing the near-infrared band of the spectrum of the remote sensing image that has a significant difference between the shadow and the non-shade. By enlarging the NSBI relative difference between the segmented landscape patch and the surrounding patches, expanding the relative spectral difference between the landscape architecture in the bright area, the landscape architecture in the shadow area, and the shadow interval, reducing the probability of spectral confusion, so that the landscape architecture and its adjacent patches can be clearly distinguished. distinguish.

The invention relates to an on-line monitoring data modeling method of bulk grid transformer equipment. The method comprises the following steps of 1) modifying substation system models, 2) allowing all-level main monitoring stations to receive the substation system models and analyzing a substation code and coordinate information in each substation system model, 3) matching the single substation system model with corresponding substation information in a primary wiring diagram of a regional grid according to the substation code and the coordinate information, 4) linking the substation system models with corresponding substation points in the primary wiring diagram of the regional grid, forming a provincial-level model file and forming version numbers, 5) forming and sending new substation system model version numbers to the all-level main monitoring stations after the substation system models are modified, and achieving automatic update of a network-level model by Step 2)-Step 4), and 6) splicing provincial-level on-line monitoring system models into a network-level on-line monitoring system model. The method is easy to realize, can reduce the update time of the model, and can raise an application level of the on-line monitoring model.

The present application relates to a method, system, device and storage medium for expanding capacity of a metadata node cluster. Multiple metadata services are deployed on each metadata node in the metadata node cluster, and each metadata service corresponds to a metadata partition. The method includes: adding a new metadata node, determining the original metadata For at least one metadata partition to be migrated out in the node, the corresponding metadata service is deployed on the newly added metadata node, and the metadata service deployed on the newly added metadata node is related to the metadata partition to be migrated out. The corresponding metadata service configuration is consistent, and the metadata partition to be migrated out is migrated to the newly added metadata node. The method can reduce the complexity of expanding the capacity of the metadata node cluster, and improve the efficiency and effect of expanding the capacity of the metadata node cluster.