38results about How to "Avoid Misidentification Situations" patented technology

The invention discloses a greenhouse intelligent management system based on big data. The greenhouse intelligent management system comprises a vision unit, an admission unit, a random verification unit, an entrance guard, an environment monitoring unit, an area division unit, a basic database, a data screening end, a rule base, a target input unit, a processor, a display unit, a storage unit and amanagement unit. The visual unit is arranged at the entrance of the greenhouse and is used for monitoring a real-time image of the entrance of the greenhouse in real time. According to the invention,the face information corresponding to the visitor is obtained from the image, and then different verification is carried out on the face information, so that the situation that the face is counterfeited by a photo or in other modes is avoided; after face verification is passed, visitors can be assessed through specified random actions, the situation that one face is mistakenly recognized is avoided, supplementary verification is conducted, meanwhile, further face discrimination can be conducted, the situation that a face is counterfeited or similar is avoided, and safety is enhanced.

The embodiment of the invention provides a leaf age identification method and device, electronic equipment and a readable storage medium, and relates to the technical field of agriculture. The method comprises the following steps: obtaining and storing initial leaf ages of plants in a target sub-region in an initial state; after the initial state, obtaining plant images of plants in the target sub-region at preset time intervals, performing core leaf recognition on the plant images, and obtaining the number of core leaves, wherein the recognized core leaves are core leaves located on main stems of the plants, and the number of the core leaves is 0 or 1; and updating the stored target historical leaf age closest to the current according to the number of the core leaves, obtaining the current leaf age of the plant is obtained, wherein after the initial state, when the plant image is obtained for the first time, the target historical leaf age is the initial leaf age. Therefore, the leaf age can be obtained through image recognition, and a large amount of manpower can be saved.

An embodiment of the invention discloses an IP address conflict processing method, equipment and a storage medium. The IP address conflict processing method comprises the steps of: acquiring host information of a target network, wherein the host information comprises a first mapping relationship between an IP address of a host in the target network and at least one host identifier; if controllinga target host corresponding to the target IP address, determining at least one host identifier corresponding to the target IP address based on the first mapping relationship; acquiring at least one host state corresponding to the at least one host identifier; and determining a target host in a controllable state based on the at least one host state, and performing control operation on the target host. Thus, when control operation is performed on the host, if the IP address conflict phenomenon occurs, the controlled target host is identified according to the host state, accurate positioning ofthe control object is realized, and the situation of host identification errors caused by IP address conflicts is avoided.

The invention discloses an information identification method, device and system. Instant messaging (IM) information is received; whether a sending subscriber belongs to a preset contact set is determined according to a subscriber identity carried in the IM information; if YES, content of the IM information is extracted, and the IM information is identified according to the content; and otherwise, the sending subscriber does not belong to the preset contact set, and the IM information is identified according to the subscriber identity. Thus, the interception effect can be improved greatly, and the information safety is improved.

The invention discloses a multi-modal intention reverse active fusion human-computer interaction method. The method comprises the steps of obtaining environment data, gesture data of a user and voicedata of the user; performing scene perception on the environment data to obtain environment information, performing gesture information extraction on the gesture data to obtain a gesture intention, and performing voice information extraction on the voice data to obtain a voice intention; performing multi-modal intention extraction on the environment information, the gesture intention and the voiceintention to obtain a fusion intention; carrying out credibility evaluation on the fusion intention to obtain a target fusion intention; and performing interactive feedback according to the target fusion intention. The fusion intention is obtained by combining multi-modal extraction of the environment moment, the gesture data and the voice data, so that intention recognition is more accurate; theold people can be prevented from repeating a certain intention due to forgetfulness; and it is determined whether the fusion intention with relatively low occurrence probability is the intention of the user or not in an active inquiry mode, and obtaining a target fusion intention reflecting the intention of the user.

The invention discloses an automatic tower crane clamp switching device and method based on material feature analysis. According to the automatic tower crane clamp switching device and method, an image acquisition device acquires an image of to-be-lifted materials located near a lifting hook, a material type recognition module analyzes the image to recognize the type and the number of to-be-lifted parts contained in the to-be-lifted materials, a hanging part specification recognition module judges specification parameters of a hanging part needed for lifting based on the type and the number of the to-be-lifted parts, a lower clamp control module controls a lower clamp, matched with the hanging part in specification parameters, on the lifting hook to clamp a clamping section, connected to the lifting hook in a sleeving mode, of the hanging part after recognizing the hanging part, and an upper clamp control module controls an upper clamp, matched with the hanging part in specification parameters, on the lifting hook to move to the clamping section and embrace the clamping section together with the lower clamp after judging that the lower clamp is clamped in place according to the clamping force. According to the device, various clamps capable of clamping the hanging parts of various models are arranged on the lifting hook, so that the lifting hook can meet the clamping requirements of the hanging parts of various models at the same time.

The invention discloses an anti-theft window with a face recognition function for security and protection. The anti-theft window comprises a protection frame, a main window and two auxiliary windows,wherein the main window is formed in one side of the protection frame, the auxiliary windows are formed in one side of the protection window, the two auxiliary windows are symmetrically arranged by taking the main window as the center, anti-theft rods are movably connected to the interiors of the auxiliary windows, movable rods are movably connected to the interior of the main window, and a rain shield is fixedly connected to the upper end of the protective frame. According to the anti-theft window with the face recognition function for security and protection, through the arrangement of the auxiliary windows, the function of transferring small goods can be achieved, after clamping blocks are separated from the anti-theft rods, the anti-theft rods can move, by rotating a handle, a windingroll is driven to rotate, a fixing rope can be tightened, the anti-theft rods can be driven to move upwards through tightening of the fixing rope, the auxiliary windows are opened, the small goods areconveniently transferred, the circulation performance is improved, the goods do not need to be conveyed from other positions, and practicability is high.

The invention discloses a metro online ticket purchasing method. The method comprises the following steps that a cellphone ticket purchasing module on an automatic ticket vending machine is selected,and a route and a payment mode are selected at the same time, wherein the automatic ticket vending machine is provided with the mobile payment ticket purchasing module, a Quick Response code scanner,a ticket delivery module and an error reporting module; the order generated by the automatic ticket vending machine is selected after purchase is carried out, payment is confirmed after the order is generated, the Quick Response code scanner scans a payment Quick Response code on a cellphone, the automatic ticket vending machine automatically charges according to the order, after charging is carried out, the automatic ticket vending machine skips into an order interface, and a Quick Response code generation request on the automatic ticket vending machine is selected; according to the selectedpayment mode, the generated Quick Response code is scanned by the cellphone to check order information recognized from the Quick Response code; after the order information is determined, the ticket delivery module on the automatic ticket vending machine is selected, if the order information is wrong, the error reporting module on the automatic ticket vending machine is selected, and after backstage workers receive the error information, manual handling is carried out in time.

The invention discloses a task flow direction identification method, which comprises the following steps: acquiring a flow chart of a target project, and determining non-flow nodes in the flow chart; wherein the non-circulation nodes are nodes in the flow chart, and the user tasks are not circulated through the nodes; constructing a residual node set according to the non-transferred nodes; setting a node where a current user task is located in the flow chart as a current node, and querying a next node of the current node from a residual node set according to a node hierarchical relationship of the flow chart; the task flow direction is determined according to the next node of the current node, so that the node where the next user task is located is determined according to the task flow direction, and the task flow direction identification efficiency and accuracy can be improved. The invention further discloses a task circulation direction identification system, a storage medium and electronic equipment, which have the above beneficial effects.

The invention discloses a commodity dynamic information generation method, device and equipment and a computer readable medium, and belongs to the technical field of computers. The method comprises the steps of collecting induction information of a goods shelf and a dynamic image of a customer; screening out an effective dynamic image corresponding to the sensing information from the dynamic images; determining first commodity information corresponding to the sensing information; determining second commodity information corresponding to the commodity in the effective dynamic image; comparing the first commodity information with the second commodity information to obtain a commodity comparison result; comparing the commodity comparison result with an information generation condition, and if the information generation condition is satisfied, generating commodity dynamic information. According to the invention, the second commodity information is obtained by using the effective dynamic image corresponding to the shelf sensing information, the second commodity information is compared with the first commodity information obtained according to the sensing information, and the commodity dynamic information is generated according to the comparison result; therefore, the condition of identification errors caused by the fact that the commodity dynamic information is generated only according to the first commodity information in the prior art is avoided.

The invention relates to a three-dimensional finger vein recognition system. The system comprises an original data acquisition module, an identification data acquisition module and a picture comparison module; the original data acquisition module comprises an original data collector and a three-dimensional model generation module; the original data collector can form any angle with a collected finger and shoot an image of a finger vein; the three-dimensional model generation module generates a three-dimensional finger vein model according to the image information shot by the original data collector at different angles, wherein the recognition data acquisition module is used for acquiring three-dimensional finger vein models of fingers and transmitting shot image information and the generated three-dimensional finger vein models to the picture comparison module, and the recognition data acquisition module is used for shooting finger vein images of the fingers put into the recognition data acquisition module and transmitting the shot finger vein images to the picture comparison module; and the picture comparison module compares the finger vein picture obtained by the identification data acquisition module with the obtained three-dimensional finger vein model, and judges whether the finger to be identified is the finger which acquires the original data or not.

The invention discloses a multi-channel test system and a channel identification method thereof, the multi-channel test system comprises a plurality of test devices and a plurality of channel modules, the channel identification method of the multi-channel test system comprises the following steps: presetting a first channel number and a second channel number, storing the first channel number in the channel module, and storing the second channel number in the channel module; storing the second channel number in the test equipment; the channel module obtains a second channel number stored in the corresponding test equipment, compares the first channel number with the second channel number, and outputs alarm prompt information if the first channel number is inconsistent with the second channel number; and the test equipment acquires the first channel number stored in the corresponding channel module, compares the first channel number with the second channel number, and controls the test equipment to stop testing if the first channel number is inconsistent with the second channel number. According to the invention, each tested battery can be bound with corresponding test equipment data one by one, so that data tracing can be carried out during later battery after-sales maintenance.