47results about How to "The counting result is accurate" patented technology

The invention discloses an aggregated white blood cell segmentation counting system. The system comprises an image acquisition module for dyeing white blood cells in a blood sample, dissolving red blood cells in the blood sample by using red blood cell lysate and acquiring a white blood cell image, an image preprocessing module used for performing image background removal on the white blood cellimage and obtaining an optimal segmentation threshold by using a maximum inter-class variance method and roughly segmenting a white blood cell region, an aggregated cell determination module used forobtaining a coarse segmentation image according to the rough segmentation of the white blood cell region, setting a discriminant function of a cell area and obtaining a multi-cell aggregation region,and an aggregated cell segmentation counting module used for extracting a cytoskeleton in each aggregation region and a gray curve at the cytoskeleton by using a morphological refinement method. According to the invention, by analyzing the gray scale characteristics of various white blood cell areas under a low power microscope, an adaptive threshold function is constructed, while a white blood cell count is obtained, the number of oxyphil cells is obtained, the cells in the aggregation region are quickly and accurately divided and counted, the method is quick and simple and is easy to implement.

The invention relates to a passenger counting system and a passenger counting method, which belong to the technical field of statistics for the number of passengers. Firstly, the passenger counting system comprises a passenger video acquisition module, a video data processing module, a video data analysis module, a passenger counting module and a boarding direction recognition module, which are connected in sequence. Secondly, the passenger counting method includes the steps: environment deployment, passenger video acquisition, video data processing, video data analysis, passenger counting and direction analysis. Under coordination of the modules, the shortage that existing computer vision and pattern recognition techniques are easy to err in case of crowdedness or poor light is overcome, and the system and the method are capable of accurately counting passengers in real time all day long under any natural conditions.

The invention relates to a target recognition method and device, a storage medium and electronic equipment, belongs to the field of computer vision and image recognition, and can recognize the total number of targets in any scene with high accuracy, thereby ensuring the accuracy of intelligent monitoring and security. The method comprises the steps of receiving an input image and identifying a scene category to which the input image belongs; And identifying and counting targets in the input image by a target counter corresponding to the identified scene category.

The invention discloses a multi-person walking race system based on a semi-active RFID electronic tag, and the system comprises an active RFID card reader, a 125KHZ low-frequency exciter, a double-frequency semi-active RFID electronic tag, a liquid crystal big screen, and an upper computer server. The 125KHZ low-frequency exciter is used for enabling a 125KHZ signal to cover a start line in a playing field, activating a 125KHZ signal of the double-frequency semi-active RFID electronic tag, and enabling the double-frequency semi-active RFID electronic tag to emit a 2.4GHZ signal. The active RFID card reader receives the 2.4GHZ signal emitted by the double-frequency semi-active RFID electronic tag. The active RFID card reader transmits the read tag signal to the upper computer server, and the upper computer server receives and processes the signals, and then transmits the processed signal to the liquid crystal big screen for displaying the number of cycles and a current place in a competition in real time. The system can achieve the full-automatic statistical operation of the scores of athletes, can display the scores of the athlete in real time, can give the scores after competition, and is accurate, instant and fair in statistical result.

The invention discloses an unmanned container system, which comprises a main body, a display component, a plurality of goods lanes and a data processing module, wherein the display component is mounted on the main body, or, is a separate display component not mounted on the main body; the plurality of goods lanes are arranged in an array mode on the inner cavity of the main body, the goods lane comprises a goods conveying device, and the goods conveying device comprises a counting component and a follower for preventing a to-be-sold goods from toppling and moving backward; and the data processing module comprises an order processing module and a payment module, the order processing module generates order information, the order information is transmitted to the payment module, and the payment module performs settlement. The unmanned container system overcomes a mistaken taking behavior which can not be overcome by the goods conveying device for unmanned vending in the prior art, the technology is simple, and the cost is low.

The invention provides a method for culturing bifidobacterium in a common incubator, a special culture medium and a corresponding count method. The special culture medium comprises an MRS-lactose solid medium and a pectin solid medium, wherein the MRS-lactose solid medium is laid on a panel bottom layer and mixed with a bacterial solution, the pectin solid medium is laid on the MRS-lactose solid medium, the panel is then inverted, and placed in the thermostatic incubator with the temperature of 37 DEG C for bacterial culturing. The special culture medium is laid by two layers, which makes theculturing of bifidobacterium in the common thermostatic incubator possible; the method for culturing bifidobacterium in a common incubator can be implemented through the common thermostatic incubator,without anaerobic incubators, anaerobic culture tanks, hungate roll-tubes and the like, and thus, investment input for equipment and medicament is low, operation is simple, and even general laboratory technicians can complete the operation after simple training.

The invention provides a jitter-free real-time rain flow counting method for structural fatigue life analysis. According to the invention, the data points which may cause the cyclic counting jitter phenomenon in the future and the corresponding cyclic values are temporarily stored by virtue of the constructed database, and once the data jitter phenomenon occurs, the cyclic counting result can be corrected, so that the cyclic counting jitter phenomenon is avoided; when the method is used for processing structural load data, the first two data points A and B are read firstly, and then a countingmethod for the data points K with odd numbers and the data points L with even numbers behind the third data point is given according to A (B or A) B in two situations. According to the method, all load process data including divergent convergence waves can be processed in real time on the premise that the sequence of the load process data does not need to be adjusted and any data points are not missed; moreover, the cyclic counting result is not affected by future data, the jitter phenomenon of cyclic counting is avoided, the counting result is more accurate, and the calculation efficiency ishigh.

An automatic counting device for measuring the number of target objects included in an image, which includes a discrimination section, a counting section, and a correction section. The discrimination section discriminates the target objects included in the image based on face information extracted from the image. The counting section measures the number of target objects discriminated by the discrimination means. The correction section corrects the number of each type of target object measured by the counting section such that the counting error arising from the difference in the discrimination accuracy between each type of target object is minimized based on information related to the discrimination accuracy for each type of target object.

The invention relates to a residual bait counting method based on computer vision. Compared with the prior art, the method comprises the following steps: 1) obtaining an underwater image, and preprocessing the image by adopting an image clarification algorithm to blur the image and lose details caused by the turbidity of the pool water; 2) considering the illumination unevenness caused by artificial light source, segmenting the foreground and background by adaptive classification and segmentation algorithm and converting it into binary image; 3) searching the contour sequence of the foregroundobject, and eliminating contours whose number of contour points does not belong to the set range; 4) removing the contour of the non-food object based on the contour feature of the foreground object;5) calculating that area of each contour after the reject treatment and sorting the area to determine the area of a single bait; 6) estimating that number of bait particle contained in the current contour and adding the estimated number of particle of all contours to obtain the final residual bait particles. The present invention has the advantages of adapting to turbidity pond water environment,adapting to uneven illumination, considering adhesion, accurate counting and the like.

The invention relates to the technical field of image recognition, and provides a cigarette filter stick recognition method and system based on machine vision, and the method comprises the following steps: collecting a vertical section image of a cigarette filter stick; the collected image is preprocessed, and a filter stick area in the image is intercepted; the filter stick is identified based on the intercepted filter stick area, the edge, the circle center and the radius of the filter stick are obtained through edge detection and circle detection, and the edge and the circle center of the filter stick are marked in the filter stick area image; carrying out abnormity elimination on the identified filter sticks; and calculating the number of circle centers and/or filter stick contours in the filter stick area image through image detection to obtain a cigarette filter stick counting result. Accurate cigarette filter stick positions and counting results can be obtained without a large amount of manual correction operation.

The invention discloses a piled profile counting system based on image recognition. The piled profile counting system specifically comprises a camera module, a marking module, a preprocessing module, a contour extraction module, a first counting module, a second counting module and an accumulation module. The method comprises the following steps: acquiring end images of piles of profiles as counting images, obtaining a binary image of the counting images, and performing curve approximation and fitting on black and white boundaries of the profiles in the binary image to obtain a first profile image capable of reflecting profile end contour features; and counting a part of profiles with complete main feature contours of the end parts in the counting image through the main feature contours, and counting the other part of profiles with local main feature contours of the end parts in the counting image through the main feature contours and the auxiliary feature contours, thereby finally realizing counting of the piles of profiles. The counting result is more accurate, counting personnel can be liberated from long-term boring transactional work, labor force is saved, and meanwhile counting efficiency is improved.