A grassland pest forecasting system

Abstract

The present invention discloses a grassland pest forecasting system in the field of ecological protection, which includes a collection module, a cloud computing module and a server. The collection module includes a drone body, and the drone body is equipped with a physical image collection module, an environmental image collection module, a level module and a flash module. A sliding rheostat is connected between the level module and the flash module. The resistance of the sliding rheostat changes according to the deflection degree of the level. Compared with the prior art that uses regional fixed-point collection, this technical solution uses the drone body equipped with the collection module to collect images, enhances the mobility of the collection process and reduces the maintenance cost of the base station in the fixed area, expands the use range of a single collection module, and is convenient for reducing the data statistics cost.

Description

Technical Field

[0001] The present invention belongs to the field of ecological protection, and specifically relates to a grassland pest forecasting system. Background Art

[0002] Pastoral areas refer to areas mainly engaged in livestock production. It is relative to agricultural areas mainly engaged in planting, forestry areas mainly engaged in forestry production, and fishing areas mainly engaged in fishery production. It is a breeding and production base for livestock and draft animals. The pastoral areas in China are mainly distributed in the west and northwest, mostly natural grasslands. Pastoral areas should highlight the important characteristics of grassland agriculture, give full play to the role of the forage industry in the grassland ecosystem, combine grass planting with livestock raising and soil cultivation, combine land with livestock, increase the diversity, stability and productivity of the grassland ecosystem, and at the same time continuously extend the ecosystem outwards, making it gradually more complex and expanded, making it more elastic, and establishing a grassland ecosystem that is mainly self-sustaining, low-input in ecology and viable in economy.

[0003] In the grassland ecosystem, pest forecasting is the basis for pest control. The existing pest surveys are still manual surveys, visually observing pest symptoms and quantities, recording survey data, mostly relying on the experience and skills of professional technicians, which is time-consuming and laborious, with a large workload. The surveyors are very hardworking, and the influence of human factors is great, far from meeting the requirements of modern forecasting. Summary of the Invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide a mechanized grassland pest forecasting system to overcome the defects of manual forecasting.

[0005] In order to achieve the above purpose, the technical solution of the present invention is as follows: A grassland pest forecasting system includes a collection module, a cloud computing module and a server.

[0006] The collection module includes a drone body, and the drone body is equipped with a physical image collection module, an environmental image collection module, a level module and a flash module. A sliding rheostat is connected between the level module and the flash module. Among them, the sliding rheostat changes the resistance according to the deflection degree of the level, and the flash module adjusts the brightness of the physical image collection module and the environmental image collection module according to the resistance change.

[0007] The cloud computing module includes a vector module, a convolution module and a screening module. The screening module is used to distinguish images with different brightness. The screening module divides the physical images and environmental images with the same brightness into a group, and then the cloud computing module transmits the grouping result to the server.

[0008] The following beneficial effects are achieved after adopting the above solution: 1. Compared with the existing technology that uses regional fixed-point collection, this technical solution uses a drone body equipped with a collection module for image collection, enhancing the mobility of the collection process and reducing the maintenance cost of base stations in fixed areas, expanding the usage range of a single collection module, and facilitating the reduction of data statistics costs.

[0009] 2. Compared with the existing technology that uses drones equipped with collection modules, this technical solution aims to overcome the defects of dynamic blurring of images taken during air flight and difficulty in screening due to the change of shooting angles of images under moving conditions, resulting in different vectors for the same scene at different angles and a relatively high repetition rate during sorting. Therefore, this technical solution uses different exposure degrees for differentiation. At this time, if the drone encounters air currents and jitters or has abnormal jitters during flight, the level deviates, thus driving the change of the resistance of the sliding rheostat.

[0010] When the resistance increases, the flash intensity decreases, and the taken pictures are dark. When the resistance decreases, the flash intensity increases, and the taken pictures are bright. The dark pictures and overexposed pictures are used to distinguish whether the drone is tilted to the left or right at this time. At the same time, when performing data processing, a threshold can be designed to directly eliminate dark pictures and overly bright pictures using the threshold, so as to screen pictures in a relatively stable state, reduce the repetition rate of the screened pictures, and improve the accuracy of image vectorization.

[0011] Furthermore, the acquisition metric module also includes a difference metric module, and the difference metric module annotates the difference in the level module for each acquired image;

[0012] The annotated physical object images and environmental images are combined according to the annotation time. Subsequently, the vector module in the cloud computing module vectorizes the combined graphics. During the vectorization process, data calculations and statistics are performed based on color depth and edge lines, and the statistically processed data is sent to the server.

[0013] Beneficial effect: Due to the volume problem of the drone, a large-volume high-definition camera cannot be carried in this technical solution. Therefore, in order to make the images clear and solve the problem of dynamic blurring, two images are input using the physical object image acquisition module and the environmental image acquisition module, one is a content image, and the other is a style image. The system outputs a synthesized image. The synthesized image has both the structural characteristics of the content image and the color and texture characteristics of the style image, so as to improve the clarity of the image and reduce the difficulty of edge detection after vectorization.

[0014] Furthermore, the server is equipped with a database that contains a gallery for analyzing the pest infestation conditions of forage grass. The server includes a matching diagnosis system and a vector matching system. The matching diagnosis system conducts a comparative analysis by comparing pictures, using photos taken at different time nodes in the same area. It vectorially compares the photo of the latest time node with the old photos, and then comprehensively judges the lack of color depth and leaf vein vectors to analyze whether pests have occurred.

[0015] Furthermore, the operating logic of the matching diagnosis system includes vector matching of leaf spot textures, vector matching of leaf vein textures, leaf stalk spot texture matching, and missing tissue vector matching.

[0016] Furthermore, the vector matching system is used to analyze the cause of pest infestation from the pathology through color depth and vector textures, so as to deduce whether pests are present and the pests that cause the pathology.

[0017] Furthermore, an alarm system is also included. The alarm system includes a warning lock, an allocation lock, and several alarm light modules. The warning lock is opened when the difference between the color depth and the vector in the matching diagnosis system exceeds the threshold, and the allocation lock is opened after the optimal matching in the database is completed to turn on the corresponding alarm light module.

[0018] Furthermore, the alarm system also includes a reserved alarm light. The reserved alarm is activated when the warning lock is opened and the allocation lock is in the closed state.

[0019] Beneficial effects: 1. Compared with the solution of photographing and distinguishing forage grass and specific pests in high-definition photos, in this technical solution, the cause of pest infestation is analyzed from the pathology through color depth and vector textures, so as to deduce whether pests are present and the pests that cause the pathology.

[0020] 2. Compared with other analysis systems, in this technical solution, combined with low-altitude drone photography, it overcomes the defect that tall forage grass covers low forage grass and the defect that leaves cover pests. The matching logic is as follows: comparative analysis is conducted using photos taken at different time nodes in the same area. The photo of the latest time node is vectorially compared with the old photos, and then the color depth and the lack of leaf vein vectors are comprehensively judged to analyze whether pests have occurred, and then the pest types are screened using the pathogenesis.

[0021] 3. Alarm is carried out through the alarm module, which is convenient for quickly determining the type of pest infestation.

[0022] Further, it also includes grassland pest image data obtained by using a camera to capture grassland pest images; calculating the histogram of the obtained grassland pest image data; performing equalization processing and broadening processing on the calculated histogram respectively; synthesizing the equalized histogram and the broadened histogram to obtain an enhanced grassland pest image; transmitting the grassland pest image to a matching diagnosis system in the server, and the matching diagnosis system also includes a pest shape database.

[0023] Beneficial effects: Through the high-definition processing of the captured images, the shapes and appearances of grassland pests are placed in the database for comparison. Brief Description of the Drawings

[0024] Figure 1 It is a schematic diagram of Embodiment 1 of the present invention;

[0025] Figure 2 It is a schematic diagram for generating a composite image in Embodiment 1. Detailed Description of the Specific Embodiment

[0026] The following is a more detailed description through specific embodiments:

[0027] Embodiment 1

[0028] Embodiment 1 is basically as shown in the attached... Figure 1 A grassland pest forecasting system includes a collection module, a cloud computing module, and a server.

[0029] The collection module includes a drone body, which is equipped with a physical image collection module, an environmental image collection module, a level module, and a flash module. There is a sliding rheostat connected between the level module and the flash module. The sliding rheostat changes its resistance according to the deflection degree of the level, and the flash module adjusts the brightness of the physical image collection module and the environmental image collection module according to the resistance change.

[0030] The cloud computing module includes a vector module, a convolution module, and a screening module. The screening module is used to distinguish images with different brightness levels. The screening module groups the physical images and environmental images with the same brightness together, and then the cloud computing module transmits the grouping results to the server.

[0031] The collection measurement module also includes a difference measurement module, and the difference measurement module marks the difference in the level module for each captured image.

[0032] The labeled physical images and environmental images are combined according to the labeling time. Then, the vector module in the cloud computing module vectorizes the combined graphics. During the vectorization process, data calculations and statistics are performed based on the color depth and edge lines, and the statistical data is sent to the server.

[0033] The server is equipped with a database, which contains a gallery of forage pest infestation condition analysis pictures. The server includes a matching diagnosis system. The matching diagnosis system conducts a comparative analysis by using pictures, comparing photos taken at different time nodes in the same area. It vectorially compares the photo at the latest time node with the old photos, and then combines the judgment of the lack of color depth and leaf vein vectors to analyze whether pests have occurred.

[0034] The operation logic of the matching diagnosis system includes leaf spot texture vector matching, leaf vein texture vector matching, leaf stalk spot texture matching, and missing tissue vector matching. The vector matching system is used to analyze the cause of pests from the pathology through color depth and vector texture, so as to deduce whether there are pests and the pests causing the pathology.

[0035] It also includes an alarm system. The alarm system includes a warning lock, a distribution lock, and several alarm light modules. The opening condition of the warning lock is that the difference between the color depth and the vector in the matching diagnosis system exceeds the threshold to open. The opening condition of the distribution lock is that after the optimal matching in the database is completed, the corresponding alarm light module is turned on. The alarm system also includes a reserved alarm light. The opening condition of the reserved alarm is that it is started when the warning lock is on and the distribution lock is off.

[0036] The specific implementation process is as follows: This technical solution uses a drone body equipped with a collection module to collect images. This technical solution uses different exposure levels for distinction. At this time, if the drone encounters air flow and shakes, or has abnormal shakes during flight, the level will deviate, which will drive the resistance of the sliding rheostat to change. When the resistance increases, the flash intensity decreases, and the taken pictures are dark. When the resistance decreases, the flash intensity increases and the taken pictures are bright. Use the dark pictures and overexposed pictures to distinguish whether the drone is tilted to the left or right at this time. At the same time, when processing data, a threshold can be designed to directly eliminate dark pictures and overly bright pictures, so as to screen pictures in a relatively stable state, reduce the repetition rate of the screened pictures, and improve the accuracy of picture vectorization.

[0037] Please refer to Figure 2 , and then use the physical image collection module and the environmental image collection module to input 2 images, one is a content image and the other is a style image. The system outputs a synthesized image. The synthesized image has both the structural characteristics of the content image and the color and texture characteristics of the style image, so as to improve the clarity of the image and reduce the difficulty of edge detection after vectorization.

[0038] Analyze the causes of pest damage from the pathology by using color depth and vector texture, so as to deduce whether there are pests and the pests causing the pathology, the defect of tall forage covering low forage, and overcome the covering of leaves on pests. The matching logic is as follows: use the photos at different time nodes in the same area for comparative analysis, vector compare the photo at the latest time node with the old photo, so as to comprehensively judge the color depth, and analyze whether there is pest damage by the absence of the vector of leaf veins. Then, screen the pest types by using the pathogenesis, and then alarm through the alarm module to facilitate quickly determining the type of pests.

[0039] Embodiment 2

[0040] The difference between this embodiment and the above embodiment is that it also includes obtaining grassland pest image data of grassland pests by using a camera; calculating the histogram of the obtained grassland pest image data; respectively performing equalization processing and broadening processing on the calculated histogram; performing synthesis processing on the histogram after equalization processing and the histogram after broadening processing to obtain an enhanced grassland pest image; transmitting the grassland pest image to the matching diagnosis system in the server, and the matching diagnosis system also includes a pest shape database.

[0041] Through the high-definition processing of the captured image, the morphology and appearance of the grassland pests are placed in the database for comparison.

[0042] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0043] The above are only embodiments of the present invention. Specific structures and characteristics and other common knowledge in the art are not described in detail herein. Those of ordinary skill in the art know all the common general technical knowledge in the technical field to which the invention pertains before the filing date or the priority date, are able to learn all the prior art in this field, and have the ability to apply conventional experimental means before this date. Those of ordinary skill in the art can, under the inspiration given in this application, combine their own abilities to improve and implement this solution. Some typical well-known structures or well-known methods should not become obstacles for those of ordinary skill in the art to implement this application. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can still be made, and these should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope claimed in this application shall be subject to the content of its claims, and the specific implementation manners and the like recorded in the specification can be used to interpret the content of the claims.

Claims

1. A grassland pest forecasting system, characterized in that: it includes a collection module, a cloud computing module and a server; The collection module includes a UAV body, which is equipped with a physical image collection module, an environmental image collection module, a level module and a flash module. A sliding rheostat is connected between the level module and the flash module. The sliding rheostat changes its resistance according to the deflection degree of the level. The flash module adjusts the brightness of the physical image collection module and the environmental image collection module according to the resistance change. When the level deflects and causes the resistance to increase, the flash intensity decreases and the captured pictures are dark; when the resistance decreases, the flash intensity increases and the captured pictures are bright. Dark pictures and overexposed pictures are used to distinguish whether the UAV is leaning to the left or right at this time; The cloud computing module includes a vector module, a convolution module and a screening module. The screening module is used to distinguish images with different brightness, directly eliminating dark pictures and overly bright pictures by using a brightness threshold, so as to screen pictures in a stable state. And the screening module groups the physical images and environmental images with the same brightness into one group, and then the cloud computing module transfers the grouping result to the server; The server has a database, and the database contains a grassland pest condition analysis picture library. The server includes a matching diagnosis system and a vector matching system. The matching diagnosis system conducts a comparison through pictures and makes a comparative analysis using pictures at different time nodes in the same area. It vectorially compares the pictures at the latest time node with the old pictures, so as to comprehensively judge the absence of color depth and leaf vein vectors to analyze whether pests have occurred; The vector matching system is used to analyze the cause of pests from the pathology through color depth and vector texture, so as to deduce whether there are pests and the pests causing the pathology. The operation logic of the matching diagnosis system includes leaf spot texture vector matching, leaf vein texture vector matching, leaf stalk spot texture matching and missing tissue vector matching; It also includes an alarm system, which includes a warning lock, an allocation lock and several alarm light modules. The opening condition of the warning lock is that the difference between the color depth and the vector in the matching diagnosis system exceeds the threshold to open, and the opening condition of the allocation lock is to turn on the corresponding alarm light module after the optimal matching in the database is completed.

Images