325results about How to "Improve counting accuracy" patented technology

A system or method for counting the number of layers of a multilayer object is adapted for counting the number of layers by means of a simple arrangement of emitting an electromagnetic wave to strike the object that is in a multilayer state. In the system or method, an electromagnetic wave is caused to strike at least either the top surface or the bottom surface of a multilayer object and electromagnetic waves generated by reflection of the incident electromagnetic wave at the respective interfaces of the layers of the multilayer object or an electromagnetic wave generated by transmission of the electromagnetic wave through the multilayer object.

A medication and/or medical supplies storage cabinet of all-steel construction has an RFID transducer or reader with an antenna array (i.e., an antenna or series of antennas) carried on a vertical elevator at the rear of the cabinet's metal shelves and drawers. The metal shelves, sides and back of the cabinet define successive compartments in which RFID-tagged item are stored. The vertical elevator may have a drive motor, e.g., gear-motor controlled by a computer associated with the cabinet. There may be a wall or window of a suitable radiolucent material at the rear of the cabinet to permit the RFID energy to radiate between the RFID antenna array and any RFID-tagged inventory items contained in the respective compartments or drawers.

The invention discloses a crowd counting method based on a coding-decoding structure multi-scale convolutional neural network, and the method is characterized in that the method comprises the following steps: considering the visual angle distortion of an image, and employing an adaptive Gaussian filter to calculate a true value density map of the image; adopting an encoding-decoding structure, building a multi-scale convolutional neural network model, wherein a loss function of the multi-scale convolutional neural network model comprises pixel space loss and counting error description; training and testing the built multi-scale convolutional neural network model to obtain a trained multi-scale convolutional neural network model; inputting a to-be-estimated image into the trained multi-scale convolutional neural network model, and predicting to obtain a crowd density map; and performing regression estimation on the crowd density map to obtain the number of people in the to-be-estimatedimage. According to the method, scale features and contextual information of the image can be reserved, and the output quality of the density map can be improved.

Pedestrian detection and counting for traffic intersection control analyzes characteristics of a field of view of a traffic detection zone to determine a location and size of a pedestrian area, and applies protocols for evaluating pixel content in the field of view to identify individual pedestrians. The location and size of a pedestrian area is determined based either on locations of vehicle and bicycle detection areas or on movement of various objects within the field of view. Automatic pedestrian speed calibration with a region of interest for pedestrian detection is accomplished using lane and other intersection markings in the field of view. Detection and counting further includes identifying a presence, volume, velocity and trajectory of pedestrians in the pedestrian area of the traffic detection zone.

The invention discloses a people flow counting method on the basis of Kinect. The method comprises the following steps of: reading in a current frame of a depth image and setting a threshold value to carry out segmentation on the depth image to obtain a threshold valve segmentation image; carrying out noise removal on the threshold valve segmentation image to obtain a tracked object image; carrying out projection processing on pixels to obtain a tracked object projection image; acquiring an outline sequence of a tracked object communication region from the projection image; acquiring a head tracked object communication region from the tracked object image to form a current frame track point set; judging tracked object affiliation of each track point and updating or newly establishing a track point set of each tracked object; judging whether the track point sets of all the tracked objects meet the tracking counting conditions and updating an entrance and exit counter; and removing the track point sets of the tracked objects who leave from a scene. The method is suitable for people flow counting at various passages, has high counting accuracy, is low in equipment complexity, is not influenced by light variation, shadow, a perspective effect and shielding of the scene and is suitable for people flow counting in different environment places.

A self-contained pill dispenser is disclosed. A housing is provided and a hopper for containing a plurality of pills is supported by the housing. A feed chute having a transport spiral therein is operatively connected to the hopper for receiving pills therefrom. The transport spiral has an outlet and controllable aperture. A vibrating plate is disposed proximate the outlet. A drive mechanism is connected to the vibrating plate for imparting vibration thereto. The drive mechanism can abruptly stop the vibration of the vibrating plate. A microcontroller is also operatively connected to the hopper, the transport spiral, and the drive mechanism. The system can self-calibrate the mechanism for counting and dispensing pills.

The invention discloses a static image crowd counting method based on joint learning, includes the following steps: pre-training stage: using 50-layer residual network to train the ImageNet 2012 classification data set to obtain the parameter initialization classifier network, classifying the image blocks into three categories by a Softmax, respectively corresponding to three regressors; In the training phase, every image block of the training data set is input to three regressors respectively, and the same image block will get different counting results. The regressor with the smallest counting error is used as a classification tag to mark the image block, and the three kinds of image blocks are used to fine-tune the regressors respectively. Classifier training phase: random sampling andensure that the number of labels of each category is consistent; Joint training phase: continuous iterative training of classifiers and regressors. The invention can carry out counting and density estimation in crowd dense scene, has certain scale self-adaptability, and improves counting precision and model generalization ability.

A tablet division feeder includes a moving unit to move a tablet T, a fixing blade located in a movement path of the tablet T, and a support plate extending from the fixing blade such that divided tablets T2 on the fixing blade are transferred and kept onto the support plate. The fixing blade divides the tablet T into upper and lower divided tablets as the tablet T is moved in such a manner that the lower divided tablet T1 is discharged and the upper divided tablet T2 is transferred from the fixing blade to the support plate by the moving unit and kept on the support plate. The upper divided tablet T2 is discharged from the support plate as the upper divided tablet T2 is further moved by the moving unit in such a manner that the upper divided tablet T2 is discharged from the moving unit.

The invention provides a novel intelligent type traffic light control system, mainly comprising a camera module, a recognizer, a server and a traffic light controller in-order cascade connection, wherein the recognizer comprises a vehicle detecting module, a flowrate detecting module, a vehicle snapshortting module, a vehicle nameplate recognition module and a vehicle model recognition module; the recognizer counts passing cars by adopting the video virtual coil technology, calculates the vehicle flow rate, recognizes the vehicle model and nameplate based on snapshortted photos and sends other information formed by the vehicle flow rate, the vehicle model and the vehicle nameplate to traffic light time operation module. The invention has the advantages of accurately counting cars, greatly increasing the amount of information acquired, being capable of working out the average waiting time of vehicle and the shortest traffic light time and saving road traffic resources.

Pedestrian detection and counting for traffic intersection control analyzes characteristics of a field of view of a traffic detection zone to determine a location and size of a pedestrian area, and applies protocols for evaluating pixel content in the field of view to identify individual pedestrians. The location and size of a pedestrian area is determined based either on locations of vehicle and bicycle detection areas or on movement of various objects within the field of view. Automatic pedestrian speed calibration with a region of interest for pedestrian detection is accomplished using lane and other intersection markings in the field of view. Detection and counting further includes identifying a presence, volume, velocity and trajectory of pedestrians in the pedestrian area of the traffic detection zone.

A self-contained pill dispenser is disclosed. A housing is provided and a hopper for containing a plurality of pills is supported by the housing. A feed chute is operatively connected to the hopper for receiving pills therefrom. The feed chute has an outlet and controllable aperture. A vibrating plate is disposed proximate the outlet. A drive mechanism is connected to the vibrating plate for imparting vibration thereto. The drive mechanism has a mechanism for abruptly stopping the vibration of the vibrating plate. A microcontroller is also operatively connected to the hopper, the feed chute, and the drive mechanism.

Pedestrian detection and counting for traffic intersection control analyzes characteristics of a field of view of a traffic detection zone to determine a location and size of a pedestrian area, and applies protocols for evaluating pixel content in the field of view to identify individual pedestrians. The location and size of a pedestrian area is determined based either on locations of vehicle and bicycle detection areas or on movement of various objects within the field of view. Automatic pedestrian speed calibration with a region of interest for pedestrian detection is accomplished using lane and other intersection markings in the field of view. Detection and counting further includes identifying a presence, volume, velocity and trajectory of pedestrians in the pedestrian area of the traffic detection zone.

The invention provides a people flow statistical system and a people flow statistical method based on video image analysis. The people flow statistical system comprises a single-image photographing unit, a speed estimating unit, a puzzle generating unit and a decision model unit, wherein the single-image photographing unit is placed just above a detection line and used for photographing a video image of the top of a pedestrian from top to bottom; the speed estimating unit is used for calculating speed information when each pedestrian passes through the detection line in the video image by using the optical flow method; the puzzle generating unit is used for calculating the distance that the pedestrian passes through the detection line in each video image according to the speed information, taking the distance as the slicing thickness, and restoring the whole puzzle of the pedestrian by continuously accumulating slices; the decision model unit is used for analyzing by utilizing linear regression according to the puzzle restored by the puzzle generating unit, calculating the number of pixels and the number of edge information in the puzzle, and counting the number of people in the puzzle. Compared with the prior art, the system and the method have the advantages that the problem that pedestrians keep out each other can be avoided, and are beneficial to improving the counting accuracy when people is crowded.

This invention provides a direct microscopic examination method milk somatic cells/bacteria automatic counter. The automatic counter comprises: a sampling/smearing apparatus, a dyeing apparatus, a washing/drying apparatus, a microscopic detecting apparatus, a convey apparatus for conveying the milk sample as well as glass slide to and out of the sampling/smearing zone, and another convey apparatus for conveying the smeared glass slide to the dyeing zone, washing/drying zone and microscopic detecting zone, which comprises multi-stage glass slide convey mechanisms and glass slide running mechanisms. The automatic analyzer comprises a smearing apparatus, a dyeing apparatus, a washing/drying apparatus a microscopic detecting apparatus and a control apparatus for convey apparatuses. This invention divides direct microscopic examination method milk somatic cells/bacteria analysis and assay into multiple mechanical actions performed by different apparatuses, and sets corresponding working positions and working zones. The working positions and working zones are connected via the convey apparatuses, and can reasonably and effectively perform automatic detection under the control of the control apparatus. The automatic counter has high counting precision and repeat precision, and the working efficiency is largely increased.

The invention discloses a solar silicon wafer and battery piece counting method based on image processing and belongs to the technical field of image processing. The method comprises the following steps that 101, side images of stacked silicon wafers or stacked battery pieces are preprocessed; 102, a measured object is positioned, and a mask is used for processing and limiting an operation region; 103, the images covered with the mask are replicated and subjected to different kinds of threshold processing respectively to obtain a denoised and binarized image; 104, the obtained denoised and binarized image is subjected to post-processing; 105, differential statistical counting and positioning are conducted to obtain the number of the measured stacked silicon wafers or the measured stacked battery pieces. According to the method, the side images of the stacked silicon wafers or the stacked battery pieces are collected, the binary image highlights gaps and filters out other interference noise, and finally a differential statistical algorithm is adopted to obtain the accurate number. By means of the method, the problem of inaccurate counting due to poor quality of the obtained image and high noise is solved, and the counting accuracy and efficiency are improved.

Paper-like counting device and trade processing device. In this counter 15, three thickness detecting sensors 20R, 20C, and 20L are juxtaposed in the vertical direction to the carrying direction of paper money 30. The thickness detecting sensors 20R, and 20L detect the thickness of both end parts of the paper money 30, and the thickness detecting sensor 20C detects the thickness of a central part of the paper money 30. The counter 15 calculates the number from the thickness of the paper money 30 detected by the thickness detecting sensor 20C. Detecting waveforms detected by the thickness detecting sensors 20R, and 20L are respectively corrected in a fold on the basis of a detecting waveform detected by the thickness detecting sensor 20C, and the number of paper money 30 is respectively calculated from the fold-corrected detecting waveforms. When the number of sheets of calculated paper money 30 wholly coincides with each other for these sensors, the number is decided to be correct number.

The invention discloses a counting device and a counting method for metallurgy bars, and belongs to the technical field of bar counting equipment. The counting device and the counting method are used for accurately counting the metallurgy bars. According to the counting device and the counting method, four sections of transmission chains are adopted to buffer bar layers, and through differential steel bar splitting and layer stepping of the transmission chains, the bars are separated one by one and stacking of the bar layers are effectively prevented. Two groups of photoelectric counting gratings mounted at different positions are utilized to count the bars, and the maximum value of counting values of the photoelectric counting gratings is taken as a counting result, so that the problem that the value of the counting result is small due to crossing of the bars is solved; influences of accumulated errors of the photoelectric counting gratings on the whole bundle of counting results are reduced by a layer correction method, and a correction method in which a small counting value is taken as a layer counting result if the small counting value of layer counting of the photoelectric counting gratings is equal to the bar number of the bar layers is adopted, so that accuracy of bar counting can be further improved. Through mutual correction of the two counters, counting accuracy is greatly improved, and the counting device has the advantages of high counting accuracy and simple structure.

The embodiment of the invention is suitable for the technical field of computers, and discloses a fluorescent cell counting method and device, terminal equipment and a computer readable storage medium. The method comprises the following steps: obtaining a to-be-counted fluorescent image which comprises at least one cell subjected to immunofluorescence staining; carrying out binarization processingon the fluorescence image to obtain a binarized image; recognizing a cell region in the binarized image according to a preset cell area parameter; and counting the number of fluorescent cells in thefluorescent image based on the cell region. According to the embodiment of the invention, the fluorescent image is directly converted into the binarized image, the cell region in the binarized image is automatically recognized, and the number of the fluorescent cells is counted based on the self-cell region, so that the counting steps are reduced, the cell region is automatically recognized and counted, the counting efficiency and counting accuracy are improved, and the operation is simple and convenient.

The invention discloses a method and a system for counting vertical reciprocating movements based on a mobile terminal. The method comprises steps: S1, a video is acquired and stored at a preset visual point; S2, the video image acquired at the current time is detected whether to have a first image associated with a vertical reciprocating movement object, if yes, the position information of the first image is acquired, and a third step is executed, and if not, the second step is returned and executed; S3, according to the position information of the first image, a rectangular frame template is adopted to calibrate a first image in each frame of video image; S4, according to a calibration time sequence, the position information of the rectangular frame template in each frame of video image is extracted, and the position information is sequentially stored in a first queue in a first queue group according to the calibration time sequence; and S5, the first queue in the first queue group is processed and analyzed to acquire whether the first image completes a preset action, if the first image completes the preset action, the times for completing the preset action are accumulated, and the third step is returned and executed.