45results about How to "Improve rejection rate" patented technology

The invention relates to an integrated circuit (IC) card payment authentication method based on fingerprints and finger veins. The method comprises the following steps: registration: a registered terminal acquires registration finger vein information and registration fingerprint information of a cardholder to generate a finger vein template and a fingerprint template, stores the fingerprint template in an IC card and the finger vein template in a background system; trading: a payment terminal obtains site payment fingerprint information and site payment finger vein information acquired when the cardholder pays at site, compares the site payment fingerprint information with the fingerprint template stored in the IC card, if the comparison fails, trading is refused, if the comparison succeeds, the background system compares the site payment finger vein information with the finger vein template stored in the background system, if the comparison fails, the trading is refused, if the comparison succeeds, the trading is authenticated. By adopting the method, the identity authentication security level can be effectively improved.

Disclosed is an online sliver separating device for high moisture content cut tobaccos. The online sliver separating device is mainly composed of a feeding vibration tank, a cut tobacco opening device, a high frequency vibration tank, a two-stage air separator and a discharging vibration tank which are sequentially arranged along the conveying direction of the cut tobaccos. Two opening devices are arranged in a first air separating channel and a second air separating channel which are communicated with each other in the two-stage air separator. Upper pin rollers and lower pin rollers are staggered and are specifically arranged on the opening devices to drop the cut tobaccos to achieve the loosening of the cut tobaccos. A separating method of the online sliver separating device includes that the high moisture content cut tobaccos which are cut in tobacco cutting production lines pass through the cut tobacco opening devices and then are subjected to high frequency vibration by means of the high frequency vibration tank to enable the cut tobaccos to be in wide cut, thin spread and fluffy states, then the cut tobaccos enter into the opening devices in the two-stage air separator, and finally slivers in the cut tobaccos are separated out. By means of the device and the method for separating the silvers, less broken cut tobaccos are produced in the separating process, and the ratio for rejecting the slivers out of the cut tobaccos is obviously increased.

The invention discloses an iris recognition system, an application method thereof and an eigenvalue extraction method for incomplete images in the iris recognition process. The iris recognition system comprises a background cloud server and one or more local iris recognition devices, wherein each local iris recognition device comprises an access controller, an access control drive valve, a local industrial personal computer and an image acquisition unit; each image acquisition unit acquires iris image data, and each local industrial personal computer recognizes the acquired iris image data by use of a stored iris recognition model and controls the access control drive valve by the aid of the access controller according to the recognition result to open access control or keep the access control closed; the local industrial personal computer in each local iris recognition device is in connection communication with the background cloud server to send sample data of the iris recognition model and receive a new iris recognition model, and the background cloud server updates the iris recognition model corresponding to each local iris recognition device on the basis of the received sample data. The iris recognition accuracy can be improved with the passage of time and the application of the system, and meanwhile, the convenience of remote system maintenance and upgrade is improved.

The invention discloses a multi-template face automatic input method in a face recognition system based on a video. The method comprises the following steps: firstly, obtaining a maximum face track ina video scene, automatically storing face pictures at multiple angles as templates according to a certain screening logic, and excluding pictures with too small faces, dynamic blurring and large angles during screening, thereby ensuring the quality of face templates; secondly, conducting illumination compensation on the template which is overexposed or insufficiently exposed, adding the compensated picture into the template of the person, and improving the comparison accuracy; and finally, converting the stored template extraction features into a face feature file, storing the face feature file into a face comparison database, and directly comparing the features, thereby reducing the operation time consumption. According to the input method, automatic collection of the user face is achieved, the input time is short, the user matching degree is low, meanwhile, the input template quality is high, and the recognition accuracy can be effectively improved.

The invention relates to an air flow and multi-stage drum type electrostatic combined foreign fiber separation device, in particular to the air flow and multi-stage drum type electrostatic combined foreign fiber separation device during unginned cotton processing. The separation device comprises a feeding port, an air flow separation device housing, an air flow separation foreign fiber inlet, an air flow opening, a foreign matter negative pressure conveying pipeline, a conveying slide plate, an electrostatic separation device housing, a motor, a roller, a pulley, a belt, an electrode plate, anelectrostatic electrode, an electrostatic separation foreign fiber inlet, a hair brush, a protective cover, a high-voltage electrostatic generator, a grounding terminal and an unginned cotton outlet;compared with the prior art, the device adopts the principle of air flow and multi-stage drum electrostatic combined sorting, and carries out air flow and multi-stage drum electrostatic combination sorting on foreign fibers, the elimination rate of the foreign fibers is higher, and the efficient separation of the unginned cotton and the foreign fibers can be achieved without damage.

The invention discloses a tobacco shred processing production line hemp rope removing device. A tobacco shred processing production line comprises one or more belt conveyers, wherein a rotating rake roller is arranged under the blanking end of at least one belt conveyer, the surface of the rake roller is provided with outward protruding rake nails, each rake nail is provided with a ball, and the diameter of each ball is larger than that of each rake nail. The rotating rake roller is arranged under the blanking end of the belt conveyer, so that when tobacco shreds fall from the blanking end of the belt conveyer, hemp ropes doped in the tobacco shreds are wound on the rake roller, and accordingly the purpose of removing the hemp ropes is achieved. The tobacco shred processing production line hemp rope removing device is high in removing rate and can thoroughly remove hemp ropes, threads and other sundries different in length. The layout of an original tobacco shred processing production line is not needed. The tobacco shred processing production line hemp rope removing device is simple in structure, low in manufacture cost, capable of greatly relieving manual labor intensity and stable in removing effect.

A delivery device of a capsule detector is disclosed, and belongs to the field of medicine container manufacturing equipment. Conventional capsule detector delivery device cannot adapt to shooting and screening of inside ends of a capsule. The delivery device comprises a driving mechanism, an inner baffle plate, an outer baffle plate, a rejecting mechanism, and vertically overlaid rotary disk and bottom disk; the driving mechanism is in transmission connection with the rotary disk; the rotary disk is provided with multiple open slots arranged in a circle, and the length of each open slot is larger than the length of a capsule; the inner baffle plate is composed of a side plate part and a separating plate part in head-to-tail connection; the side plate part is arc and stretches across the open slots; the distance between the bottom surface of the side plate part and the top surface of the bottom disk is less than the diameter of a capsule, and the distance between the inner wall of the side plate part and the inner end of each open slot is equal to the length of a capsule; the separating plate part divides the rotary disk into a material falling zone and a shooting zone; the outer baffle plate is arranged at the lateral side of the rotary disk in the shooting zone; the finishing end of the shooting zone is provided with a capsule channel; the inlet end of the separating plate part is provided with a slipping through hole; the outer baffle plate is provided with an outer end shooting opening; and a vent of the rejecting mechanism is at the bottom disk under the open slots. The delivery device helps to realize shooting and screening on the inner ends of a capsule.

The invention discloses a gesture recognition method based on a regional full convolutional network. The method comprises: for an input gesture image, performing feature extraction through a full convolutional network to obtain a set of feature maps and generate a candidate box, a position sensitive sub-network generating a position sensitive score map, and scoring each gesture category through apooling layer, so that positioning and classification of target gestures are realized. The method is mainly characterized in that the whole regional full convolutional network is of a shared full convolutional structure. The whole structure is end-to-end learning, complex calculation is avoided while the high-precision recognition rate is achieved, the OHEM technology is combined. The network model has the higher rejection rate on negative samples, practical application is facilitated, and the method has important significance in the field of human-computer interaction.

The invention relates to ginsenoside and an ultrasonic extraction method thereof. The extraction method comprises the steps of sample treatment, ultrasonic extraction, silica gel adsorption, eluent decolorization, concentration drying and the like. The extraction method is short in extraction time, low in extraction temperature and high in extraction efficiency. Besides, an ethanol solution is adopted as a solvent in the whole extraction and separation process, other toxic organic solvents are not used, and environmental pollution is small while the extraction efficiency is guaranteed. According to the method, a chromatographic separation technology is adopted, and impurities in the extract are thoroughly removed, so that the quality of the ginsenoside product is obviously improved.

The invention relates to a cigarette case packaging detection process. The cigarette case packaging detection process sequentially comprises the steps of conveying, detecting, removing and collectingand is realized based on a cigarette case packaging detection system. The cigarette case packaging detection system is sequentially provided with a paper feeding mechanism, a detection mechanism, a removing mechanism and a collecting mechanism in the conveying direction of a conveying belt. To-be-detected packaging paper is conveyed to the detection mechanism at a uniform speed piece by piece at intervals through the paper feeding mechanism. Good products or defective products are detected through the detection mechanism. The good products enter the rear collecting mechanism directly through agood product channel. The defective products are introduced into a defective product channel to enter the rear collecting mechanism through the removing mechanism. According to the cigarette case packaging detection process, an integrated production process is formed through the conveying and detection of the to-be-detected packaging paper and the separating, conveying and collecting of the goodproducts and the defective products, and detecting automation is realized. All working procedures are tightly matched through a device, so that the detection efficiency and the detection accuracy areimproved, and the labor intensity is reduced.

The embodiment of the invention discloses an impurity recognition and removal device for agricultural stem-leaf products, and a working method thereof. The impurity recognition and removal device comprises a feeding conveyor belt, an ultraviolet light detecting imaging device, a feeding slideway, a discharging conveyor belt and a programmable logic controller (PLC), wherein the ultraviolet light detecting imaging device is arranged above the feeding conveyor belt; an ultraviolet light emitted by the ultraviolet light detecting imaging device can completely shine on a material on the feeding conveyor belt; the ultraviolet light detecting imaging device is connected with the PLC through a computer image processor; the feeding slideway is arranged inclinedly; the feeding conveyor belt and thedischarging conveyor belt are connected to two ends of the feeding slideway; a blanking device is arranged on the feeding slideway and is connected with the PLC; and the PLC controls the blanking device to remove impurities in the material. The impurity recognition and removal device for the agricultural stem-leaf products, and the working method thereof provided by the invention solve the problems of narrow application range, high technology difficulty, and non-ideal impurity removal effect of an existing impurity removal device for the agricultural stem-leaf products.

The invention provides multi-channel mildewed tea seed kernel sorting equipment, and belongs to the technical field of tea seed oil processing. The defect that mildewed tea seed kernels are prone to being mixed in tea seed oil processing in the prior art is overcome. The multi-channel mildewed tea seed kernel sorting equipment comprises a vibration disc, first arrangement grooves, second arrangement grooves and a controller, the upper ends of the first arrangement grooves are fixedly connected with an outlet of the vibration disc, the first arrangement grooves and the second arrangement grooves are aligned in a one to one mode, a fixing frame is fixedly arranged on each of the first arrangement grooves and the second arrangement grooves, a rotating mechanism is arranged between the first arrangement grooves and the second arrangement grooves, and a set of mildewed tea seed kernel screening system is arranged on each of the first arrangement grooves and the second arrangement grooves. The mildewed tea seed kernel screening system comprises a mildewing detection mechanism for judging whether tea seeds are mildewed or not according to the color of tea seed kernels and a high-speed pneumatic sorting mechanism for removing the mildewed tea seed kernels. According to the equipment, the removal rate of the mildewed tea seed kernels can be improved, and improving of the finished product quality of tea seed oil is facilitated.

The invention discloses a cigarette empty end detection system, which comprises a power supply, a synchronization unit, a data acquisition unit, an analog-to-digital conversion unit, a main control unit, and a rejection unit; the output terminal of the synchronization unit and the input of the analog-to-digital conversion unit and the rejection control unit Both terminals are connected, the synchronization unit is used to provide a synchronous clock signal for the analog-to-digital conversion unit, the main control unit and the rejection control unit, the output terminal of the capacitive sensor is connected to the input terminal of the analog-to-digital conversion unit, and the output terminal of the analog-to-digital conversion unit The input end of the main control unit is connected, and the output end of the main control unit is connected with the rejecting unit; a kind of cigarette end detection system of the present invention adopts the capacitive sensor with good stability and can adapt to various complex environments as the detection device, The hardware circuit adopts programmable control FPGA as the main processing chip, which not only can meet the complex and changeable production environment of the tobacco factory, but also improves the rejection rate of problematic cigarettes and greatly improves the quality of the product.

The invention belongs to the technical field of artificial reproduction of fish fry, and discloses a method for removing female pseudobagrus fulvidraco and male pseudobagrus fulvidraco hybrid offspring deformed fry. The method comprises the following steps that 1, fry are temporarily raised in a cement pool; 2, micro flowing water is kept in the temporary raising pool, an uninterrupted inflation state is kept, and temporary raising is carried out for 70-75 hours; 3, the fry which are scattered, not clustered and float after gradually dying are directly removed, the fry in a clustered state are taken out through tulle or siphon or the fry are directly centralized in a fry box from a drainage opening to be taken out, and a fry culture stage is executed. According to the method, the female pseudobagrus fulvidraco and male pseudobagrus fulvidraco hybrid offspring deformed fry are removed, operation is easy, the requirement on equipment is low, and the removal rate is high.

The invention relates to a full-automatic product detection machine. The full-automatic product detection machine is successively provided with a paper feeding mechanism (1), a detection mechanism (2), an eliminating mechanism (3) and a collecting mechanism (4) in the conveying direction of a conveyor belt; wrapping paper to be detected is alternately conveyed into the detection mechanism (2) at aconstant speed piece by piece through the feeding mechanism (1), good products or defective products are detected by the detection mechanism (2), the good products directly enter the rear collectingmechanism (4) through a good product channel, and the defective products are imported into a defective product channel by the eliminating mechanism (3) and enter the rear collecting mechanism (4). According to the full-automatic product detection machine, from collection of convey of the wrapping paper to be detected, detection and separate convey and collection of the good products and defectiveproducts, an integrated technological process is formed, detection automation is realized, each working procedure is closely matched through the full-automatic product detection machine, the detectionefficiency and the detection accuracy are improved, and labor intensity is lowered.