43 results about "Insect identification" patented technology

The invention discloses an automatic early warning system for pests and diseases comprising a server and a pest situation real-time measuring system; the server comprises a human-computer interaction interface, a data analysis module, an insect identification and classification module and an image storage module; the pest situation real-time measuring system comprises a central processing unit and a communication module, wherein the central processing unit is connected with an image acquisition unit, an image processing unit and a mechanical operation unit. The system of the invention has following beneficial effects: the system is practical in that it can classify and count killed insets as well as kill the insects; insect recognition rate is high; the system is easy to use; the system can monitor in real time and save labor cost, and is green and environmentally friendly. The problems that collecting field pest type and quantity by hand is difficult and data lacks real-time performance due to agricultural production characters of large plantation area, numerous in variety and complicated production environment are solved. Therefore, the automation and real-time of plant protection platform are fulfilled.

The invention discloses a remote insect identification and counting device based on machine vision. The device comprises a solar power supply device, an insect trapping and counting device, an image acquisition chamber, an insect body recovery device, an environment detection device and a wireless transmission device. The insect trapping and counting device comprises a waterproof top cover, a trapping chamber, an insect trapping agent tray, an insect trapping lamp source, a contact killing power grid and a current detection module; the image acquisition chamber comprises an insect body conveying belt and a miniature camera unit; the insect body recovery device comprises a turnover baffle and a recovery bottle; the environment monitoring device comprises a luminous intensity sensor, an airtemperature and humidity sensor and a soil humidity sensor; the wireless transmission device comprises a microcomputer, a storage unit and a 4G transmission module. The insect image can be acquired and analyzed in real time and remotely, besides, insects are counted, and the intelligent level of insect monitoring in the agricultural district is improved.

The invention provides a light trap insect identification counting method and a system thereof, wherein the method and the system relate to the field of agricultural plant protection technology. The method comprises the steps of trapping insects by means of an arranged insect trapping lamp so that the insects stop on an insect trapping board; acquiring an insect image on the insect trapping board by an image acquisition unit every first preset time length, transmitting the insect image to a server; segmenting the acquired insect image to insect sub-images which respectively comprise a single insect by a server; inputting the insect sub-image into a trained deep learning identification model for realizing identification classification; and performing statistics on the kind and the number of the insects which appear in the insect images acquired in a second preset time length every second preset time length. Compared with the prior art, the light trap insect identification counting method and the system thereof are advantageous in that full automation of insect trapping, automatic identification and counting is realized through matching of the insect trapping lamp, the insect trapping board, the image acquisition unit and the server; insect identification and statistics efficiency are improved; and furthermore insect identification efficiency is improved.

The invention provides a diptera insect recognition method based on a deep convolutional neural network, and the method comprises the following steps: collecting a diptera insect image, and making a data set of the diptera insect image; performing data enhancement on the data set; constructing an improved RetinaNet target detection model; setting training parameters, and training the RetinaNet target detection model through the data set; and classifying and positioning the test set images of the diptera insects based on the trained target detection model. According to the diptera insect identification method based on a deep convolutional neural network, the RetinaNet target detection model is adopted, meanwhile, an improved convolution block attention module is added, a feature pyramid network is improved, a large amount of manpower and material resources do not need to be consumed, the problem of dependence on manual design features can also be solved, and the image acquisition methodis easy to operate.

The invention provides an insect identification early warning method and system, and relates to the technical field of agricultural plant protection. According to the method and the system, various insect images are collected through an image collection apparatus; the obtained insect images are input to a trained deep learning identification model for performing identification classification to obtain insect types and corresponding insect states; the insect states are judged; and when the insect states meet a preset condition, an early warning prompt is sent out. The method and the system can automatically identify the insect types and the corresponding insect states, so that the insect identification accuracy and efficiency are improved; and in addition, the early warning prompt facilitates related personnel to early discover suitable dates for insect prevention and control.

The invention provides an insect identification device and method based on three-dimensional simulation. The device comprises a standard posture insect three-dimensional reconstruction module, an insect three-dimensional information acquisition module and an insect identification module, wherein the standard posture insect three-dimensional reconstruction module is used for constructing a standard posture insect three-dimensional model library; the insect three-dimensional information acquisition module is used for obtaining the image, the three-dimensional posture and the two-dimensional feature information of an insect to be identified; the insect identification module is used for carrying out the primary election of insect species on the basis of the three-dimensional posture and the two-dimensional feature information of the insect to be identified, a corresponding three-dimensional model is selected according to the primary election insect species in the three-dimensional model library, deformation is carried out according to the three-dimensional posture information to obtain the three-dimensional model of the corresponding posture of each type of insect and independently obtain a two-dimensional projection image corresponding to the three-dimensional model, the two-dimensional image feature information of the insect to be identified is calculated, a Mahalanobis distance between the obtained two-dimensional projection image feature information is obtained, and the insect type corresponding to the two-dimensional projection image corresponding to a minimum Mahalanobis distance is taken as the type of the insect to be identified. By use of the method and the device, an insect identification accuracy rate can be improved.

The invention provides an agricultural pest identification method based on a convolutional neural network, a terminal and a readable storage medium. The method comprises the steps of collecting insectimage information; setting a label of the insect image information, and performing feature extraction to obtain feature mapping information; judging state information of the feature region information; mapping the candidate region to a ROIs region with a fixed size obtained on the feature map, and judging a target category; inputting the test set into a training model, and detecting insect information in the candidate region pictures; and when the precision of the detected candidate area picture reaches a predetermined value, completing the establishment of the training model, and detecting insect information in the image through the training model. According to the invention, the identification degree of insect image information is improved, and the detection precision of insect identification is improved. By effectively utilizing the agricultural pest identification method based on the convolutional neural network, pest images can be identified in agricultural pest image identification, so that the monitoring capability of agricultural automation is improved.

The invention relates to an insect identification and counting system, and the system comprises a scattering identification and counting system for recording scattered light signals triggered by the vibration of the body and wings of the insect, a shading identification and counting system for recording shading light signals triggered by the vibration of the wings of the insect, a signal acquisition processor for triggering the counting of the insect through the time delay of the shading light signals relative to the scattered light signals and judging the insect type through the scattered light signals and the shading light signals, and a terminal processing display for displaying the counting result of the insect and the insect type. The scattering identification and counting system is positioned above the shading identification and counting system. The invention also relates to a field insect monitoring instrument which comprises an attracting system, a capturing system and an insect identifying and counting system. The capturing system is positioned below the shading identification and counting system. The insect identification and counting system has the advantages of simple structure and accurate counting, integrates insect trapping, optical identification and optical counting technologies, and belongs to the technical field of insect monitoring.

The invention belongs to the technical field of computer teaching software, and discloses a virtual teaching system and a virtual teaching method for amber embedded specimen experimental teaching. Thesystem comprises an anatomy acquisition module, an image preprocessing module, a picture export module, a three-dimensional modeling module, a model export module, a map model import module, an interactive production module, a virtual reality module and an information release module. The virtual teaching system comprises an anatomy acquisition module, an image preprocessing module, a picture export module, a three-dimensional modeling module, a model export module, a map model import module, an interactive production module, a virtual reality module and an information release module. The present invention is useful for insects from living environment, morphological characteristics, cameras and video cameras are used to collect images and make videos of insect behaviors and physiological structures into libraries, through the three-dimensional form of insect identification and identification, in the experimental classroom teaching to show to students, can not only improve the teachingeffect, help students better grasp the knowledge of insects, but also lay a solid foundation for field practice.

The invention provides a system and a method for identifying varieties of insects based on a big data cloud server. The problems that the existing insect identification cannot meet the requirements and the automatic identification process is complex are solved. The identification system comprises a mobile terminal and the big data cloud server; the mobile terminal comprises a picture acquiring module with close-distance photography, a display module for displaying a picture acquired by the picture acquiring module, a GPS module with a positioning function and a data transmission module for transmitting the picture acquired by the picture acquiring module; the big data cloud server comprises an identification server for classifying insects, an insect information database for storing insectdetailed information and a cloud management module; the cloud management module is connected with the insect information database; and the data transmission module of the mobile terminal is connectedwith the identification server through wireless network. Meanwhile, the invention also discloses the related insect identification method. By the system and the method, unknown insects can be identified rapidly, and convenience is provided for people to prevent injurious insects and utilize beneficial insects.

The invention relates to a grain insect identification method and identification system based on SSD algorithm feature fusion. The identification method comprises the following steps: establishing a data set; establishing a neural network model and training the neural network model by adopting the data set to obtain a trained neural network model; and collecting a to-be-identified grain insect image, inputting the to-be-identified grain insect image into the trained neural network model, and detecting the type and position of the grain insect in the to-be-identified grain insect image. According to the technical scheme provided by the invention, the output feature maps of the convolution layers conv43 and conv53 in the neural network model are fused, the block 11 unfavorable for small target detection is deleted, and the priori box suitable for the grain insect is obtained by adopting the Kmeans clustering algorithm, so that the defect of the default priori box in the original SSD is improved, the identification and positioning of the grain insect are facilitated, the problem that in the prior art, grain insect recognition accuracy is poor can be solved.

The invention relates to a carmine spider mite SCAR (sequence characterized amplified region) mark and a specificity PCR (polymerase chain reaction) detection method thereof, belonging to the field of plant insect identification technologies. The carmine spider mite SCAR mark has the nucleotide sequence as shown in Seq (sequence) ID (identity) No.2; and the specificity PCR detection method of carmine spider mite is characterized in that specificity forward and reverse primers adopted in a PCR amplification system are designed by the nucleotide sequence of the SCAR mark described in claim 1, and an amplification region of the specificity forward and reverse primers is arranged on the nucleotide sequence of the SCAR mark. Compared with the traditional identification method, the method is simple, fast, time-saving, labor-saving and easy to master for staff in a classification major and the non-classification majors, thereby being the molecular biological method which can fast identify the carmine spider mite.

The invention discloses a method and system for identifying aphids on yellow boards. The identification method comprises the steps of obtaining a yellow board image in a cotton field environment; performing adaptive threshold segmentation on the yellow board image twice, and determining a segmented yellow board image; determining a yellow board insect image according to the yellow board image andthe segmented yellow board image; obtaining insect types corresponding to the historical insect morphological characteristics; establishing an insect recognition network model by taking the historicalinsect morphological characteristics as input and taking the insect species as output; obtaining insect morphological characteristics of the yellow plate insect image; according to the insect morphological characteristics and an insect identification network model, identifying the insect species in the cotton large field, and determining aphids on the yellow plate; and according to the aphids onthe yellow plate, determining the quantity of the aphids in the cotton large field. According to the method and the system for identifying the aphids on the yellow board, the accuracy and the automation of aphid identification are improved.

Identifying insect species integrates image processing, feature selection, unsupervised clustering, and a support vector machine (SVM) learning algorithm for classification. Results with a total of 101 mosquito specimens spread across nine different vector carrying species demonstrate high accuracy in species identification. When implemented as a smart-phone application, the latency and energy consumption were minimal. The currently manual process of species identification and recording can be sped up, while also minimizing the ensuing cognitive workload of personnel. Citizens at large can use the system in their own homes for self-awareness and share insect identification data with public health agencies.

The invention discloses an insect flower visiting behavior analysis method based on a machine learning model, which relates to the technical field of machine learning, and comprises the following steps: acquiring insect flower visiting videos of set types of flowering plants through video acquisition equipment within a set time period, and according to the types, distribution regions and floweringseasons of the flowering plants, selecting a corresponding insect identification model from a pre-created insect identification model set, inputting the insect flower visiting video into the insect identification model to obtain the type of each insect in each frame of image of the insect flower visiting video, and according to the type of each insect in each frame of image of the insect flower visiting video, and obtaining a flower visiting behavior analysis result of each insect on the same kind of flowering plants in the same time period. The above steps are repeated to obtain the flower visiting behavior analysis results of the insects on the flowering plants of the same kind in the same time period, the flower visiting behavior analysis results of the insects on the flowering plantsof different time periods, different regions and different kinds are obtained, the insect kind identification accuracy is improved, the insect flower visiting behavior analysis result accuracy is guaranteed, and the expandability is expanded.

The present invention provides an insect identification device and method based on three-dimensional attitude estimation. The device comprises a three-dimensional attitude information acquisition module for acquiring three-dimensional attitude information of an insect to be identified, and the three-dimensional attitude information includes insect wing rotation angles and trunk deformation information. The device also comprises an insect feature extraction module and an insect type identification module. The insect feature extraction module is used for obtaining feature information of the insect to be identified, wherein the feature information comprises a form feature, a color feature and a texture feature. The insect type identification module is used for a pre-established SVM-based insect identification module to identifying the insect to be identified according to the three-dimensional attitude information and the feature information, so as to obtain the type identification result of the insect to be identified. The insect identification rate can be effectively improved.

The invention relates to a method for quickly counting the number of pests, comprising the steps of: vertical photography, area selection, insect identification and quantity statistics, mainly throughthe use of cameras or mobile phones to perpendicularly photograph the pests lying on the wall or in containers to obtain electronic photographs, and then from the electronic photographs to select theidentification area of evenly distributed pests, and then to identify the species of pests in the identification area and quantity statistics. The method can be used to take photos and identify quickly by mobile phone conveniently, and the accuracy rate of the number of pests is higher than 85%. The method has obvious advantages of fast speed and high accuracy, and is suitable for the preventionand control of diseases and pests in granary and the research experiment of pests.

It is described an insect attack prediction system, comprising: at least one processor provided with a plurality of software modules comprising: an insect identification module configured to process at least one insect digital image (IM) to provide a presence value (IPD), representing the presence of insects in an area of interest for insect attack; a data collecting module configured to acquire insect behavioral data associated to said area and comprising at least one of the following data groups: meteorological data; environmental data; historical data of insect presence. The system further comprises a prediction module configured to process the presence value (IPD) and the insect behavioral data according to a mathematical prediction algorithm to estimate a risk of attack (PRB) to the area of interests.

Identifying insect species integrates image processing, feature selection, unsupervised clustering, and a support vector machine (SVM) learning algorithm for classification. Results with a total of 101 mosquito specimens spread across nine different vector carrying species demonstrate high accuracy in species identification. When implemented as a smart-phone application, the latency and energy consumption were minimal. The currently manual process of species identification and recording can be sped up, while also minimizing the ensuing cognitive workload of personnel. Citizens at large can use the system in their own homes for self-awareness and share insect identification data with public health agencies.

The invention discloses a multi-resolution visual identification method for grain insect species. The method comprises steps as follows: three shot grain insect images with low resolution, medium resolution and high resolution are binarized with an OTSU algorithm respectively, connected regions of the grain insect images with low resolution, medium resolution and high resolution are extracted witha Blob algorithm, and a suspected grain insect target area is positioned in the grain insect image with low resolution; reverse positioning is performed according to magnification of the images withmedium resolution and low resolution and the suspected grain insect target area in the grain insect image with low resolution, area parameters and perimeter parameters are calculated, and whether thesuspected grain insect target area is a grain insect area is determined; finally, on the basis of the connected region, extracted with the Blob algorithm, of the image with high resolution, reverse positioning is performed according to magnification of the images with high resolution and medium resolution and the determined medium-resolution grain insect area, and grain insect identification is realized by a random forest classifier according to LBP (local binary pattern) grain insect texture characteristics. The grain insect species are identified efficiently and accurately by the aid of advantages of development tools and image processing algorithms.

The invention discloses a visual Transform-based small sample insect image recognition method, which comprises the following steps of: firstly, searching images of various insects by utilizing a search engine, and manually labeling the images with labels; then, constructing a pre-training model taking a visual Transform as a core, and carrying out optimization training on parameters in the pre-training model by utilizing the training set; removing a classifier in the pre-training model, randomly extracting a small number of image samples of each type of insects in the training set and the test set, inputting the image samples into a visual Transform to extract image features, and calculating an average value of each type of samples as prototype features of each type of insect images to be stored; and finally, acquiring insect images on line, inputting the insect images into Transformer to extract image features, calculating the distance between the insect images and each type of insect image representation, and outputting the insect type with the closest distance as the type of the image. According to the method, the insects are classified and identified by using a small number of training samples, and the technical problem that a large number of image samples are needed during convolutional neural network training used in current insect identification can be solved.

The invention belongs to the technical field of insect identification. The invention discloses an insect identification method based on three-dimensional simulation, which is provided with a processorbox body, a camera photographing module is fixed in the middle of the processor box body through screws, a sound collecting module is installed beside the camera photographing module through screws,and a glass cover is installed on the surface of the processor box body through a rotating shaft. The camera shooting module is connected with a picture data memory, the picture data memory is connected with a picture binary processor, and the picture binary processor is connected with a picture feature sampler. The sound acquisition module is connected with the sound data memory, and the sound data memory is connected with the sound characteristic sampler. The picture feature sampler is coupled to a central processing unit which transfers data to a data comparison analyzer together with the sound feature sampler. By collecting and comparing the appearance pictures and sound information, we can accurately analyze the insect species, so that ordinary people can quickly grasp the important information of the insect.

The invention discloses a dragoneye insect identification method based on regional suggestion network, which comprises the following steps of: S1, cleaning and sorting dragonproof insect images to obtain a data set of the dragonproof insect images; S2, performing enhancement processing on the data set of the dragoneye insect image to obtain an enhanced data set; S3, dividing the enhanced data set to obtain a training set, a verification set and a test set of dragoneye insect images; S4, constructing a deep convolutional network model based on the regional suggestion network; S5, training a deep convolutional network model by using the training set and the verification set to obtain a trained network model; and S6, inputting the test set into the trained network model, and outputting and obtaining a classification result of the test set. According to the method, the identification processing is simple and fast, a large amount of labor cost is saved, and the problem of difficult identification caused by complex background of dragoneye insect pictures in a natural environment is solved.

Provided in the present disclosure are an insect identification method and system. The system comprises a client and a server, and the client is connected to the server. The method comprises: the client acquires an image photograph that is captured by a target user and that contains an insect to be identified, acquires an insect region in the image photograph, slices the insect region to obtain an insect image, and sends the insect image to the server; the server identifies the insect image by invoking an insect species identification model deployed in the server, and sends the identification result to the client; and the client pushes the identification result to the target user. By applying the solution provided in the present disclosure, the problem in the prior art in which insect identification is very difficult and has low accuracy may be solved.

The invention discloses an insect identification workbench which comprises a table, a first microscope and a second microscope are arranged at the top end of the table, baffles are fixed to the left side, the right side and the front side of the top end of the table, and a cold light source lamp is installed on the baffle located on the front side through a clamping mechanism; the first screw rod is driven to rotate through the rotating handle, so that the frame body on the first screw rod moves up and down along the first screw rod, the height of the table can be properly adjusted according to insect researchers with different heights, the working comfort of the insect researchers is improved, and convenience and comfort are mainly embodied; by forming the slide clamping groove, the culture dish containing groove, the tool containing groove and the alcohol bottle groove in the table, space is saved, and multiple steps of dissecting and identifying insects are achieved; common specimens and slide specimens can be identified at the same time through the design of the two microscopes, so that comfortable operation is achieved to the maximum extent, and the insect identification workbench can be popularized and used in observation and identification personnel of insects and other arthropods.

The invention relates to the field of insect identification, in particular to a specific gene and a molecular identification method of culicolides brevipalpis. The specific gene is one of a CO I geneand a Cyt b gene, or a combination of the two; the sequence of the CO I gene is SEQ ID NO. 1, the sequence of the Cyt b gene is SEQ ID NO. 2, and the specific gene can be used for accurate identification of the species of the culicolides brevipalpis. The invention provides the molecular identification method of the culicolides brevipalpis, which combines the CO I gene with the Cyt b gene and has high accuracy and short identification time; during homology comparison among DNAs of to-be-detected insects and the CO I and Cyt b genes of the culicolides brevipalpis, the to-be-detected insects arejudged as the culicolides brevipalpis when the homology reaches 98% or above.

The invention relates to the technical field of insect research and discloses an insect identification system, which includes a module carrier, an observation cup, a storage battery, a ring side wall, a top wall, a bottom wall, a hole, a ring slide, a slide chamber, a hole, a slide Blocks, connecting blocks, wheels, rubber strips, card holes, card balls, slots and light-tight inserts, capture components include square tubes, brackets, insect nets, card slots one, crossbars, rods one, rod two, rods 3. Ring, card slot 2, card protrusion 2, card protrusion 1, annular elastic ring, jack, cover plate, vent hole, control module includes microchip controller, mobile phone client, cloud server, wheel drive module, image A processing module, a wireless transmission module, a camera and an image memory; the invention also discloses an identification method of an insect identification system. The identification result of the present invention is accurate and reliable, saves more time and energy of the user, realizes semi-intelligent control when luring insects, saves time and labor, and can obtain high-definition images of insects at 360° angles.