Plant identification method, plant identification device and plant identification system

Abstract

The present disclosure relates to a plant recognition method, a plant recognition device, and a plant recognition system. The plant recognition method includes: when the image acquisition unit is in an operating state, detecting image acquisition parameters based on a preview image generated by the image acquisition unit; automatically adjusting the image acquisition parameters according to the detection result and / or sending a prompt message for instructing a user to adjust the image acquisition parameters; obtaining a captured image from the image acquisition unit based on the adjusted image acquisition parameters, wherein the captured image includes an image of a plant to be recognized; and recognizing the plant to be recognized according to the captured image.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a plant identification method, a plant identification device and a plant identification system. Background Art

[0002] In order to identify plants, users can provide plant images, and these images are identified to obtain information such as the plant species. However, since there may be various problems with the plant images, the accuracy and efficiency of identification may be very low. Therefore, there is a need for improved plant identification methods, devices and systems. Summary of the invention

[0003] One of the purposes of the present disclosure is to provide a plant identification method, a plant identification device and a plant identification system.

[0004] According to a first aspect of the present disclosure, a plant identification method is proposed, comprising:

[0005] When the image acquisition unit is in operation, detecting image acquisition parameters according to the preview image generated by the image acquisition unit;

[0006] Automatically adjusting image acquisition parameters according to the detection results and / or issuing prompt information for instructing the user to adjust the image acquisition parameters;

[0007] Based on the adjusted image acquisition parameters, acquiring a captured image from the image acquisition unit, wherein the captured image includes an image of a plant to be identified; and

[0008] The plant to be identified is identified according to the captured image.

[0009] In some embodiments, the image acquisition parameters include at least one of the following: the plant region where the plant in the preview image is located, the shooting distance of the plant, the shooting focal length of the plant, the number of plants, and the position of the plant.

[0010] In some embodiments, detecting the image acquisition parameters according to the preview image generated by the image acquisition unit includes:

[0011] A plant region recognition model is used to determine the plant region where the plant in the preview image is located.

[0012] In some embodiments, detecting the image acquisition parameters according to the preview image generated by the image acquisition unit includes:

[0013] Detecting a plant region where the plant in the preview image is located, and determining an area of ​​the plant region;

[0014] Calculating the area ratio of the region area to the total area of ​​the preview image; and

[0015] The shooting distance of the plant is determined according to the area ratio.

[0016] In some embodiments, detecting a plant region where the plant in the preview image is located and determining the area of ​​the plant region includes:

[0017] Determine a marked frame circumscribed to the plant, and determine the area of ​​the marked frame as the area of ​​the plant region; or

[0018] Using a Mask RCNN model to determine a Mask region corresponding to the plant region, and determining the area of ​​the Mask region as the area of ​​the plant region; or

[0019] A multi-target recognition model is used to respectively determine the plant region where each plant in the preview image is located, and to respectively determine the area of ​​each plant region.

[0020] In some embodiments, automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes:

[0021] Determining whether the area ratio is within a preset ratio range; and

[0022] When the area ratio is not within the preset ratio range, at least one of the following operations is performed until the area ratio is within the preset ratio range:

[0023] Automatically adjust the focus of your shots; and

[0024] A first prompt message is issued to instruct the user to adjust the shooting distance and / or the shooting focal length.

[0025] In some embodiments, automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes:

[0026] Determining whether the area ratio is within a preset ratio range;

[0027] When the area ratio is not within the preset ratio range, automatically adjusting the shooting focal length until the area ratio is within the preset ratio range, or the shooting focal length reaches a maximum focal length or a minimum focal length; and

[0028] When the shooting focal length reaches the maximum focal length or the minimum focal length, and the area ratio is not within the preset ratio range, a second prompt message is issued to instruct the user to adjust the shooting distance until the area ratio is within the preset ratio range.

[0029] In some embodiments, the plant identification method further comprises:

[0030] When the area ratio is within the preset ratio range, at least one of the following prompt messages is issued:

[0031] Third prompt information for indicating to the user that the shooting focal length and shooting distance are appropriate; and

[0032] The fourth prompt information is used to instruct the user to stop adjusting the shooting focal length and the shooting distance.

[0033] In some embodiments, automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes:

[0034] When it is detected that the number of plants in the preview image is more than one, fifth prompt information is issued to instruct the user to select at least one of the plants as the plant to be identified;

[0035] receiving a selection from a user, and determining a plant to be identified in the preview image according to the selection; and

[0036] An area including at least a portion of the plant to be identified is determined as a focus area in the preview image.

[0037] In some embodiments, automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters further includes:

[0038] When it is detected that the number of plants in the preview image is more than one, a labeling frame circumscribing each of the plurality of plants is used to label the corresponding plants respectively.

[0039] In some embodiments, detecting the image acquisition parameters according to the preview image generated by the image acquisition unit includes:

[0040] A plant part recognition model is used to determine the plant part in the preview image.

[0041] In some embodiments, automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes:

[0042] When it is detected that the plant to be identified in the preview image includes a strong characteristic part, the area including the strong characteristic part is determined as the focus area in the preview image.

[0043] In some embodiments, detecting the image acquisition parameters according to the preview image generated by the image acquisition unit includes:

[0044] According to a preset priority order, the part of the plant to be identified that has the highest priority and is contained in the preview image is determined as the strong feature part.

[0045] In some embodiments, automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes:

[0046] Identifying a preliminary classification of the plant to be identified in the preview image;

[0047] Acquiring strong characteristic part information of the plant to be identified according to the preliminary classification; and

[0048] An area containing the strong feature part of the plant to be identified and corresponding to the strong feature part information is determined as a focus area in the preview image.

[0049] In some embodiments, identifying the preliminary classification of the plant to be identified in the preview image includes:

[0050] A pre-recognition classification model is used to identify the preliminary classification of the plant to be identified in the preview image.

[0051] In some embodiments, obtaining the strong characteristic part information of the plant to be identified according to the preliminary classification includes:

[0052] According to the preliminary classification, the strong characteristic part information of the plant to be identified is retrieved from a first preset database.

[0053] In some embodiments, automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes:

[0054] Identifying a preliminary classification of the plant to be identified in the preview image;

[0055] Determining whether the preliminary classification belongs to the easily confused category;

[0056] When the preliminary classification of the plant to be identified belongs to an easily confused category, determining the distinguishing characteristic part information of the plant to be identified according to the easily confused category; and

[0057] An area including the distinguishing characteristic part of the plant to be identified and corresponding to the distinguishing characteristic part information is determined as a focus area in the preview image.

[0058] In some embodiments, determining whether the preliminary classification belongs to the easily confused category includes:

[0059] According to the preliminary classification, a search is performed from a second preset database to determine whether there is a confusing category corresponding to the preliminary classification.

[0060] In some embodiments, determining the distinguishing characteristic part information of the plant to be identified according to the easily confused category includes:

[0061] According to the easily confused category, the distinguishing characteristic part information of the plant to be identified is retrieved from a second preset database.

[0062] In some embodiments, based on the adjusted image acquisition parameters, acquiring the captured image from the image acquisition unit includes:

[0063] The image acquisition unit automatically focuses on the determined focus area; and

[0064] The image acquisition unit performs photographing to obtain a photographed image.

[0065] In some embodiments, automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes:

[0066] When it is detected that the preview image does not include plants, sixth prompt information is issued to instruct the user to adjust the shooting range.

[0067] In some embodiments, identifying the plant to be identified according to the captured image includes:

[0068] A plant recognition model is used to identify the plant to be identified based on the captured image.

[0069] According to a second aspect of the present disclosure, a plant identification device is proposed, including a processor and a memory, wherein the memory stores instructions, and when the instructions are executed by the processor, the steps of the plant identification method described above are implemented.

[0070] According to a third aspect of the present disclosure, a plant identification system is proposed, comprising:

[0071] An image acquisition unit, wherein the image acquisition unit is configured to generate a preview image and acquire a captured image;

[0072] an image detection unit, wherein the image detection unit is configured to detect image acquisition parameters according to the preview image generated by the image acquisition unit;

[0073] at least one of a parameter adjustment unit and a user prompt unit, wherein the parameter adjustment unit is configured to automatically adjust the image acquisition parameters according to the detection result, and the user prompt unit is configured to issue prompt information for instructing the user to adjust the image acquisition parameters according to the detection result; and

[0074] A plant identification unit is configured to identify the plant to be identified based on the captured image.

[0075] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which instructions are stored. When the instructions are executed, the steps of the plant identification method described above are implemented.

[0076] According to a fifth aspect of the present disclosure, a computer program product is proposed. The computer program product includes instructions. When the instructions are executed by the processor, the steps of the plant identification method described above are implemented.

[0077] Other features and advantages of the present disclosure will become more apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings, which constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0079] The present disclosure may be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

[0080] Figure 1 A schematic diagram showing a network environment according to an exemplary embodiment of the present disclosure is shown;

[0081] Figure 2 A schematic diagram showing a process flow of a plant identification method according to an exemplary embodiment of the present disclosure is shown;

[0082] Figure 3 A schematic flow chart of step S110 of a plant identification method according to a specific example of the present disclosure is shown;

[0083] Figure 4 A schematic flow chart of step S120 of a plant identification method according to a specific example of the present disclosure is shown;

[0084] Figure 5A schematic flow chart of step S120 of a plant identification method according to another specific example of the present disclosure is shown;

[0085] Figure 6 A schematic flow chart of step S120 of a plant identification method according to another specific example of the present disclosure is shown;

[0086] Figure 7 A schematic flow chart of step S120 of a plant identification method according to another specific example of the present disclosure is shown;

[0087] Figure 8 A schematic flow chart of step S120 of a plant identification method according to a further specific example of the present disclosure is shown;

[0088] Fig. 9 A schematic flow chart of step S130 of a plant identification method according to a specific example of the present disclosure is shown;

[0089] Fig.10 A schematic diagram of a plant identification device according to an exemplary embodiment of the present disclosure is shown;

[0090] Fig.11 A schematic diagram of a plant identification system according to an exemplary embodiment of the present disclosure is shown.

[0091] Note that in the embodiments described below, sometimes the same reference numerals are used in common between different drawings to represent the same parts or parts with the same functions, and their repeated descriptions are omitted. In some cases, similar numbers and letters are used to represent similar items, so once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0092] For ease of understanding, the position, size, range, etc. of each structure shown in the drawings and the like may not represent the actual position, size, range, etc. Therefore, the present disclosure is not limited to the position, size, range, etc. disclosed in the drawings and the like. DETAILED DESCRIPTION

[0093] Various exemplary embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure.

[0094] The following description of at least one exemplary embodiment is in fact merely illustrative and is in no way intended to limit the present disclosure and its application or use. That is, the structures and methods herein are shown in an exemplary manner to illustrate different embodiments of the structures and methods in the present disclosure. However, those skilled in the art will appreciate that they merely illustrate exemplary ways of the present disclosure that can be implemented, rather than exhaustive ways. In addition, the drawings need not be drawn to scale, and some features may be enlarged to illustrate the details of specific components.

[0095] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.

[0096] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0097] Figure 1 A schematic diagram of a network environment 200 according to an exemplary embodiment of the present disclosure is shown. The network environment 200 may include a mobile device 202, a remote server 203, a training device 204, and a database 205, which are coupled to each other by wire or wirelessly via a network 206. The network 206 may be embodied as a wide area network (such as a mobile phone network, a public switched telephone network, a satellite network, the Internet, etc.), a local area network (such as Wi-Fi, Wi-Max, ZigBee™, Bluetooth™, etc.), and / or other forms of networking functions.

[0098] The mobile device 202 can be a mobile phone, a tablet computer, a laptop computer, a personal digital assistant and / or other computing devices configured to collect, store and / or transmit images such as digital photos. Therefore, the mobile device 202 may include an image acquisition unit such as a digital camera and / or may be configured to receive images from other devices. The mobile device 202 may include a display. The display may be configured to provide one or more user interfaces to the user 201, and the user interface may include multiple interface elements, and the user 201 may interact with the interface elements, etc. For example, the user 201 may use the mobile device 202 to take pictures of plants and upload or store images. The mobile device 202 may output information about the species of the plant to the user and recommend maintenance guidelines suitable for the plant, etc.

[0099] The remote server 203 may be configured to analyze plant images received from the mobile device 202 via the network 206 to determine the type of the plant and recommend maintenance guidelines, etc. The remote server 203 may also be configured to create and train a plant recognition model, a plant region recognition model, a plant part recognition model, a pre-recognition classification model, etc. as described below.

[0100] The training device 204 can be coupled to the network 206 to facilitate the training of plant recognition models, plant region recognition models, plant part recognition models, pre-recognition classification models, etc. The training device 204 can have multiple CPUs and / or GPUs to assist in training plant recognition models, plant region recognition models, plant part recognition models, pre-recognition classification models, etc.

[0101] The following will take the training process of the plant recognition model as an example to explain the specific model training process, and the training of other models can be similarly performed. In one embodiment, the plant recognition model can be established based on training a neural network, and its training process is as follows:

[0102] A certain number of image samples annotated with corresponding information are obtained for each plant species, and the number of image samples prepared for each plant species may be equal or unequal. The corresponding information annotated for each image sample may include the plant name in the image sample (including scientific name, alias, category name of botanical classification, etc.). The image samples obtained for each plant species may include images of the plant of the species at different angles, different lighting conditions, different seasons (for example, the morphology of the same plant may be different in different seasons), different times (for example, the morphology of the same plant may be different in the morning and evening of each day), different growth environments (for example, the morphology of the same plant may be different when grown indoors and outdoors), and different geographical locations (for example, the morphology of the same plant may be different when grown in different geographical locations). In these cases, the corresponding information annotated for each image sample may also include information such as the angle, lighting, season, time, growth environment, and geographical location of the image sample.

[0103] The image samples that have undergone the above-mentioned annotation processing are divided into a training sample set for training the plant recognition model and a test sample set for testing the training results. Usually, the number of samples in the training sample set is significantly larger than the number of samples in the test sample set. For example, the number of samples in the test sample set accounts for 5% to 20% of the total number of image samples, and the number of samples in the corresponding training sample set accounts for 80% to 95% of the total number of image samples. It should be understood by those skilled in the art that the number of samples in the training sample set and the test sample set can be adjusted as needed.

[0104] The neural network is trained using the training sample set, and the output accuracy of the trained neural network is tested using the test sample set. If the output accuracy does not meet the requirements, the number of image samples in the training sample set is increased, and the neural network is retrained using the updated training sample set until the output accuracy of the trained neural network meets the requirements. If the output accuracy meets the requirements, the training ends. In this way, the trained neural network with an output accuracy that meets the requirements can be used as a trained plant recognition model.

[0105] The neural network may include, for example, a deep convolutional neural network (CNN) or a deep residual network (Resnet). The deep convolutional neural network is a deep feedforward neural network, which uses a convolution kernel to scan a plant image, extracts features to be identified in the plant image, and then identifies the features to be identified in the plant. In addition, in the process of identifying a plant image, the original plant image can be directly input into the deep convolutional neural network model without preprocessing the plant image. Compared with other recognition models, the deep convolutional neural network model has higher recognition accuracy and recognition efficiency. Compared with the deep convolutional neural network model, the deep residual network model adds an identity mapping layer, which can avoid the phenomenon that the accuracy of the convolutional neural network is saturated or even decreased as the network depth (the number of stacked layers in the network) increases. The identity mapping function of the identity mapping layer in the residual network model needs to satisfy: the sum of the identity mapping function and the input of the residual network model is equal to the output of the residual network model. After the introduction of the identity mapping, the change of the output of the residual network model is more obvious, so the recognition accuracy and recognition efficiency of the plant physiological period recognition can be greatly improved, thereby improving the recognition accuracy and recognition efficiency of the plant.

[0106] It should be noted that the concepts of the present disclosure may be practiced using other known or future developed training and recognition models.

[0107] The database 205 can be coupled to the network 206 and provide the data required by the remote server 203 for related calculations. For example, the database 205 may include a first preset database for storing information on strong characteristic parts of plants, a second preset database for storing the names of easily confused categories of plants and information on distinguishing characteristic parts of easily confused categories (in some embodiments, the second preset database may also store shooting rules corresponding to these distinguishing characteristic parts), etc. The database 205 may also include a material database containing materials such as wallpapers. For example, a large number of wallpaper images may be stored in the material database. After the type of plant is identified, the type may be associated with the corresponding wallpaper stored in the material database for user use or collection. The database may be implemented using various database technologies known in the art. The remote server 203 may access the database 205 as needed to perform related operations.

[0108] It should be understood that the network environment in this article is only an example. Those skilled in the art can add more devices or delete some devices as needed, and can modify the functions and configurations of some devices.

[0109] In order to improve the image quality of the captured images of plants used for identification, and thus improve the accuracy and efficiency of plant identification, the present disclosure proposes a plant identification method. Figure 2 As shown, the plant identification method may include:

[0110] Step S110: when the image acquisition unit is in operation, image acquisition parameters are detected according to the preview image generated by the image acquisition unit.

[0111] The image acquisition unit may include but is not limited to independently arranged cameras, video cameras, cameras included in mobile devices such as smart phones and tablet computers, as well as image acquisition cards, video acquisition cards, etc.

[0112] When the image acquisition unit is in operation, it can generate a preview image and display the preview image on a display screen of a camera, a camcorder, a mobile device or a computer, etc. Based on the preview image, the user can know the image that will be obtained when shooting in the current state, so it is convenient to observe and adjust during the shooting process.

[0113] In order to improve the accuracy of identifying the photographed plants, the image acquisition parameters can be detected according to the preview image (it should be noted that, usually, the captured image of the plant corresponding to the preview image has not yet been obtained at this time), and the image acquisition parameters can be adjusted to an appropriate range in subsequent steps to improve the quality of the final captured image of the plant.

[0114] The image acquisition parameters may include at least one of the plant region where the plant in the preview image is located, the shooting distance of the plant, the shooting focal length of the plant, the number of plants, and the position of the plant. It is understandable that in some other embodiments, the image acquisition parameters may also include other specific parameters.

[0115] In the process of detecting the image acquisition parameters, if the preview image includes multiple plants, a plant to be identified may be first selected from the multiple plants, and then at least one of the plant area, shooting distance, shooting focal length, and part of the plant to be identified may be detected. Alternatively, at least one of the plant area, shooting distance, shooting focal length, and part of each plant may be detected respectively.

[0116] In some embodiments, a plant region recognition model may be used to determine the plant region where the plant in the preview image is located. For example, a labeling box circumscribed to the plant may be determined, and the region where the labeling box is located may be used as the plant region. Alternatively, a Mask RCNN model may be used to determine the Mask region, and the Mask region may be used as the plant region. Alternatively, a multi-target recognition model may be used to respectively determine the plant region where each plant in the preview image is located.

[0117] In some embodiments, Figure 3 As shown, the image acquisition parameters are detected according to the preview image generated by the image acquisition unit. Specifically, the shooting distance of the plant may include:

[0118] Step S111a, detecting the plant region where the plant in the preview image is located, and determining the area of ​​the plant region;

[0119] Step S111b, calculating the area ratio of the region area to the total area of ​​the preview image; and

[0120] Step S111c, determining the shooting distance of the plant according to the area ratio.

[0121] Among them, when the plant area is represented by a labeling box circumscribed with the plant, the area of ​​the labeling box can be determined as the regional area of ​​the plant area. In addition, for the convenience of processing, the labeling box can generally be rectangular. Alternatively, when the plant area is a Mask area determined by the Mask RCNN model or the plant area includes multiple plant areas determined by the multi-target recognition model, the area of ​​each plant area can also be calculated. It can be understood that, when other conditions remain unchanged, when the area ratio of the regional area to the total area of ​​the preview image is larger, it can be determined that the shooting distance of the plant is closer, and when the area ratio of the regional area to the total area of ​​the preview image is smaller, it can be determined that the shooting distance of the plant is farther.

[0122] In some embodiments, detecting image acquisition parameters according to the preview image generated by the image acquisition unit, specifically detecting plant parts, thereby helping to determine strong characteristic parts and / or distinguishing characteristic parts of the plant in subsequent steps may include:

[0123] A plant part recognition model based on, for example, neural network training is used to determine the plant parts in the preview image.

[0124] Among them, the strong characteristic parts refer to the parts of the plant with significant characteristics, and different parts may have different priorities according to the significance of the characteristics of each part. Among them, the characteristics of the parts with higher priorities are more significant. For example, the characteristic parts of the plant may include the flower part, fruit part, leaf part, stem part and bud part of the plant. For some plants, the flower part and fruit part of the plant may have a first priority, the leaf part of the plant may have a second priority, and the stem part and bud part of the plant may have a third priority, wherein the first priority is higher than the second priority, and the second priority is higher than the third priority. It is understandable that the priorities of various parts may be different depending on the plant species. For example, if the leaf part of a certain plant has the most significant characteristics, then the leaf part of the plant may have the highest first priority, while the priorities of other parts are lower.

[0125] The distinguishing characteristic parts of a plant refer to the parts that can be used to distinguish different plants belonging to easily confused categories. In a specific example, a group of easily confused categories of plants include peach blossoms and cherry blossoms, and the corresponding distinguishing characteristic parts are the flower parts of the plant. Among them, peach blossoms are melon seed-shaped, with a point at the outer end of the petals; while the outer end of the petals of cherry blossoms has a triangular notch. In another specific example, a group of easily confused categories of plants include roses and roses, and the corresponding distinguishing characteristic parts are the stem parts of the plant. Among them, the stems of roses are smooth, and the tops of the thorns have a slight curve; while the stems of roses have more thorns, and the tops of the thorns are not bent, and there are some small hairs on the stems.

[0126] By detecting the parts of the plants in the preview image and including the strong characteristic parts and / or distinguishing characteristic parts of the plant to be identified in the final focus area as much as possible, the accuracy and efficiency of plant identification can be improved, as will be described in further detail later.

[0127] return Figure 2 , the plant identification method may also include:

[0128] Step S120: automatically adjusting image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters.

[0129] Specifically, when it is detected that the image acquisition parameters corresponding to the current preview image are inappropriate, for example, the shooting distance is too large or too small, the currently focused plant or the focused part of the plant cannot well reflect the plant to be identified or the significant features of the plant to be identified, etc., the image acquisition parameters can be automatically adjusted, or a prompt message can be issued to instruct the user to adjust the image acquisition parameters according to the prompt message, thereby improving the recognition effect of the plant.

[0130] In some embodiments, Figure 4 As shown, automatically adjusting the image acquisition parameters according to the detection results and / or issuing prompt information for instructing the user to adjust the image acquisition parameters may include:

[0131] Step S121a, determining whether the area ratio is within a preset ratio range; and

[0132] Step S121b, when the area ratio is not within the preset ratio range, perform at least one of the operations of automatically adjusting the shooting focal length and issuing a first prompt information for instructing the user to adjust the shooting distance and / or the shooting focal length until the area ratio is within the preset ratio range.

[0133] Specifically, in some embodiments, when the area ratio is less than the minimum preset ratio, it indicates that the plant is too small in the preview image. At this time, the shooting focal length can be automatically increased, or a first prompt message for instructing the user to reduce the shooting distance and / or increase the shooting focal length is issued. After receiving the first prompt message, the user can approach the plant to reduce the shooting distance or manually adjust the shooting focal length, so that the plant area accounts for a larger proportion in the preview image, which helps to obtain a clearer image of the plant. In some embodiments, when the area ratio is greater than the maximum preset ratio, it may be difficult to obtain the full picture of the plant in the preview image or to help accurately identify the characteristic parts of the plant. Therefore, the shooting focal length can be automatically reduced, or a first prompt message for instructing the user to increase the shooting distance and / or reduce the shooting focal length is issued. After receiving the first prompt message, the user can move away from the plant to increase the shooting distance or manually adjust the shooting focal length, so that more parts of the plant are displayed in the preview image, thereby improving the recognition effect based on a more complete image of the plant. It can be understood that in different embodiments, within the preset ratio range can be defined as an area ratio greater than or equal to a minimum preset ratio, or within the preset ratio range can be defined as an area ratio less than or equal to a maximum preset ratio, or within the preset ratio range can be defined as an area ratio greater than or equal to the minimum preset ratio and less than or equal to the maximum preset ratio.

[0134] In some embodiments, Figure 5 As shown, automatically adjusting the image acquisition parameters according to the detection results and / or issuing prompt information for instructing the user to adjust the image acquisition parameters may include:

[0135] Step S122a, determining whether the area ratio is within a preset ratio range;

[0136] Step S122b, when the area ratio is not within the preset ratio range, automatically adjusting the shooting focal length until the area ratio is within the preset ratio range, or the shooting focal length reaches the maximum focal length or the minimum focal length; and

[0137] Step S122c, when the shooting focal length reaches the maximum focal length or the minimum focal length, and the area ratio is not within the preset ratio range, a second prompt message is issued to instruct the user to adjust the shooting distance until the area ratio is within the preset ratio range.

[0138] In this embodiment, in order to reduce the operations that need to be performed by the user and improve the user experience, you can first try to adjust the area ratio to a suitable preset ratio range by automatically adjusting the shooting focal length. When the area ratio cannot be adjusted to the preset ratio range by automatically adjusting the shooting focal length, a second prompt message for instructing the user to adjust the shooting distance is issued to prompt the user to adjust the area ratio according to the second prompt message. The adjustment range of the focal length can be determined according to the performance parameters of the image acquisition unit, etc., to avoid blurring of the final captured image.

[0139] Specifically, in some embodiments, when the area ratio is less than the minimum preset ratio, it indicates that the plant is too small in the preview image. At this time, the shooting focal length can be automatically increased first until the area ratio is within the preset ratio range. If the shooting focal length has been increased to the maximum focal length, but the area ratio is still less than the minimum preset ratio, a second prompt information for instructing the user to reduce the shooting distance can be issued, and the user can approach the plant according to the second prompt information so that the plant area accounts for a larger proportion in the preview image, thereby helping to obtain a clearer image of the plant. In some embodiments, when the area ratio is greater than the maximum preset ratio, it may be difficult to obtain the full picture of the plant in the preview image or to accurately identify the characteristic parts of the plant. Therefore, the shooting focal length can be automatically reduced first until the area ratio is within the preset ratio range. If the shooting focal length has been reduced to the minimum focal length, but the area ratio is still greater than the maximum preset ratio, a second prompt information for instructing the user to increase the shooting distance can be issued, and the user can move away from the plant according to the second prompt information so that the part of the plant displayed in the preview image can be larger, thereby improving the recognition effect based on a more complete image of the plant. Similarly, in different embodiments, within the preset ratio range can be defined as an area ratio greater than or equal to a minimum preset ratio, or within the preset ratio range can be defined as an area ratio less than or equal to a maximum preset ratio, or within the preset ratio range can be defined as an area ratio greater than or equal to the minimum preset ratio and less than or equal to the maximum preset ratio.

[0140] It should be noted that, in order to prevent the area ratio from being adjusted too large or too small, when it is detected that the area ratio has been adjusted to be within the preset ratio range, the automatic change of the shooting focal length can be stopped immediately.

[0141] Moreover, when the area ratio is adjusted to be within a preset ratio range, a third prompt message may be issued to instruct the user that the shooting focal length and shooting distance are appropriate, and / or a fourth prompt message may be issued to instruct the user to stop adjusting the shooting focal length and shooting distance, thereby promptly reminding the user that the current shooting focal length and shooting distance are appropriate and no further adjustments are required to avoid excessive adjustment of image acquisition parameters.

[0142] In some embodiments, Figure 6 As shown, automatically adjusting the image acquisition parameters according to the detection results and / or issuing prompt information for instructing the user to adjust the image acquisition parameters may include:

[0143] Step S123a, when it is detected that the number of plants in the preview image is more than one, issuing fifth prompt information for instructing the user to select at least one of the plants as the plant to be identified; and

[0144] Step S123b, receiving a selection from the user, and determining the plant to be identified in the preview image according to the selection; and

[0145] Step S123c: determining a region including at least a portion of the plant to be identified as a focus region in the preview image.

[0146] To improve the accuracy and efficiency of recognition, when the preview image includes more than one plant, a fifth prompt message may be issued to remind the user to select one or more plants that he or she wants to identify, and the focus area in the preview image may be determined based on the user's selection.

[0147] In some embodiments, the plant to be identified selected by the user may be one. In this case, other image acquisition parameters may be determined based on the plant to be identified, without determining corresponding image acquisition parameters for other plants in the preview image, thereby helping to significantly improve the efficiency of identification.

[0148] In other embodiments, the number of plants to be identified selected by the user may also be multiple. When the preview image includes multiple plants to be identified, during the identification process, an area that only includes at least a portion of one plant to be identified may be determined as a focus area each time, and the focus area may be focused, photographed, and identified, and then the above operations may be performed for the next plant to be identified. Alternatively, during the identification process, an area that includes at least a portion of at least two or all of the multiple plants to be identified may be determined as a focus area, that is, in the focus area, relevant information about at least two or all of the plants to be identified may be obtained, and the focus area may be focused, photographed, and identified, so that multiple plants may be identified at one time. In particular, when multiple plants to be identified are close to each other and their high-quality images can be obtained simultaneously using the same or similar image acquisition parameters, such a one-time identification will help improve the recognition efficiency while ensuring the accuracy of the recognition.

[0149] Furthermore, at least a portion of the plant to be identified that is included in the focus area may contain a strong characteristic portion and / or a distinguishing characteristic portion of the plant to be identified, as will be explained in detail below.

[0150] In some embodiments, in order to facilitate the user to identify and select the plant to be identified from multiple plants in the preview image, automatically adjusting the image acquisition parameters according to the detection results and / or issuing prompt information for instructing the user to adjust the image acquisition parameters may also include:

[0151] When it is detected that the number of plants in the preview image is more than one, the corresponding plants are respectively labeled using a labeling frame circumscribing each of the multiple plants.

[0152] It is understandable that in some other embodiments, other methods may be used to represent multiple plants in the preview image to facilitate user identification and selection, such as by outputting a corresponding plant list, etc., which is not limited here.

[0153] In some embodiments, automatically adjusting image acquisition parameters according to the detection results and / or issuing prompt information to instruct the user to adjust the image acquisition parameters may include redetermining the focus area of ​​the preview image according to the detection results, so that the focus area contains more and / or more important information for identifying plants, thereby improving the plant identification effect.

[0154] For example, automatically adjusting the image acquisition parameters according to the detection results and / or issuing prompt information for instructing the user to adjust the image acquisition parameters may include:

[0155] When it is detected that the plant to be identified in the preview image includes a strong characteristic part, the area containing the strong characteristic part is determined as the focus area in the preview image.

[0156] As mentioned above, the strong characteristic parts refer to the parts of the plant with significant characteristics. Therefore, the type of the plant and other information can be identified more accurately and efficiently based on the strong characteristic parts of the plant.

[0157] In addition, in some embodiments, the parts of the plant to be identified that have the highest priority and are included in the preview image can be determined as strong feature parts in the order of preset priorities. For example, when the flower part and the fruit part of the plant have the first priority, the leaf part of the plant has the second priority, and the stem part and the bud part of the plant have the third priority, and the first priority is higher than the second priority, and the second priority is higher than the third priority, it is detected whether the plant to be identified in the preview image includes the flower part or the fruit part. If so, the flower part or the fruit part is included in the focus area. If not, it is continued to detect whether the plant to be identified in the preview image includes the leaf part. If so, the leaf part is included in the focus area. If not, it is continued to detect whether the plant to be identified in the preview image includes the stem part or the bud part. If so, the stem part or the bud part can be included in the focus area.

[0158] As described above, the strong feature parts with the highest priority may be different in different plants, so the plants to be identified can be preliminarily classified, and the strong feature parts to be included in the focus area can be determined based on the results of the preliminary classification. Figure 7 As shown, automatically adjusting the image acquisition parameters according to the detection results and / or issuing prompt information for instructing the user to adjust the image acquisition parameters may include:

[0159] Step S124a, identifying the preliminary classification of the plant to be identified in the preview image;

[0160] Step S124b, obtaining strong characteristic part information of the plant to be identified according to the preliminary classification; and

[0161] Step S124c: determining a region including a strong feature part of the plant to be identified and corresponding to the strong feature part information as a focus region in the preview image.

[0162] In some embodiments, a pre-recognition classification model based on, for example, neural network training may be used to identify a preliminary classification of the plant to be identified in the preview image.

[0163] Specifically, in some embodiments, identifying the preliminary classification of the plant to be identified in the preview image may include: acquiring and recording one or more of the location information and season information when the preview image is acquired, excluding impossible plant classifications according to one or more of the location information and season information; and, in the case of excluding impossible plant classifications, using a pre-identification classification model to identify the preliminary classification of the plant. For example, the possibility that the plant is a palm tree can be excluded based on the fact that the preview image of a certain plant is generated in Northeast China, because in China, palm trees are usually only distributed in areas south of the Qinling Mountains except Tibet. For example, the possibility that the plant is a pear flower can be excluded based on the fact that the preview image of the plant is generated in winter, because pear flowers usually only bloom in spring.

[0164] In other embodiments, the preliminary classification of the plant to be identified in the preview image may include: identifying the genus information of the plant. For example, after the above-mentioned preliminary classification and identification process of the plant, it may only be possible to identify the genus information of the plant (for example, peach, cherry or rose, etc.) but not accurately identify its species information (i.e., its accurate category). For example, it can only be identified that the plant belongs to the genus peach, but it cannot be determined which peach species it is. In this embodiment, the strong characteristic parts of the plant can be determined based on the pre-established correspondence between the genus of the plant and its corresponding characteristic parts. For example, for the peach plant, the fruit, petal morphology, calyx, overall morphology (for example, whether it is a tree or a shrub), whether the branches are hairy or not, or whether the front and back of the leaves are hairy or not, etc., can be further judged to further confirm the accurate category of the peach plant. For plants of the genus Sakura, the accurate category of the plant can be further determined by judging whether the calyx is reflexed, whether the calyx is hairy, the length of the sepals and the calyx tube, the overall shape of the inflorescence, the bracts, the overall shape of the leaves, whether the leaves are hairless on both sides, whether the leaf edges are serrated, the shape of the petal tops, and the shape of the stipules. For plants of the genus Rose, the accurate category of the plant can be further determined by judging whether the calyx is reflexed, whether the calyx is hairy, the length of the sepals and the calyx tube, the overall shape of the inflorescence, the bracts, the overall shape of the leaves, whether the leaves are hairless on both sides, whether the leaf edges are serrated, the shape of the petal tops, the shape of the stipules, and whether the flower stems have thorns or the shape of the thorns. Based on this, the strong characteristic parts corresponding to the peach plants can be pre-established to include one or more of the fruit, petals, calyx, whole body, branches and leaf parts, etc.; the strong characteristic parts corresponding to the cherry plants can be pre-established to include one or more of the calyx, sepals, calyx tube, petals, bracts and leaf parts, etc.; the strong characteristic parts corresponding to the rose plants can be pre-established to include one or more of the calyx, sepals, calyx tube, petals, bracts, leaves and flower stem parts, etc.

[0165] The corresponding relationship between the preliminary classification of the above-mentioned plants and the strong feature parts (i.e., strong feature part information) can be stored in the first preset database. Correspondingly, obtaining the strong feature part information of the plant to be recognized according to the preliminary classification may include:

[0166] Retrieving the strong feature part information of the plant to be recognized from the first preset database according to the preliminary classification.

[0167] In some embodiments, when it is determined according to the preliminary classification of the plant that the plant belongs to an easily confused category, in order to improve the recognition accuracy and efficiency, the distinguishing feature parts to be included in the focused area can be determined according to the preliminary classification of the plant. Specifically, as Figure 8 shown, automatically adjusting the image acquisition parameters according to the detection result and / or sending a prompt message for instructing the user to adjust the image acquisition parameters may include:

[0168] Step S125a, identifying the preliminary classification of the plant to be recognized in the preview image;

[0169] Step S125b, determining whether the preliminary classification belongs to an easily confused category;

[0170] Step S125c, when the preliminary classification of the plant to be recognized belongs to an easily confused category, determining the distinguishing feature part information of the plant to be recognized according to the easily confused category; and

[0171] Step S125d, determining the area containing the distinguishing feature part corresponding to the distinguishing feature part information of the plant to be recognized as the focused area in the preview image.

[0172] Similarly, the plants in the easily confused category and their corresponding distinguishing features can be stored in the second preset database. Correspondingly, determining whether the preliminary classification belongs to an easily confused category may include: retrieving from the second preset database whether there is an easily confused category corresponding to the preliminary classification according to the preliminary classification. In addition, determining the distinguishing feature part information of the plant to be recognized according to the easily confused category may include: retrieving the distinguishing feature part information of the plant to be recognized from the second preset database according to the easily confused category.

[0173] In some embodiments, considering that the user may not have aligned any plant for shooting, in order to prompt the user for this situation to improve the recognition effect, automatically adjusting the image acquisition parameters according to the detection result and / or sending a prompt message for instructing the user to adjust the image acquisition parameters may further include:

[0174] When it is detected that the preview image does not include a plant, sending a sixth prompt message for instructing the user to adjust the shooting range.

[0175] When receiving the sixth prompt message, the user can recheck the shooting distance, range and other parameters to include at least a part of the plant to be identified in the preview image, and further continue to adjust the image acquisition parameters accordingly based on the preview image containing the plant to be identified, so as to obtain better recognition effect.

[0176] return Figure 2 , the plant identification method may also include:

[0177] Step S130: acquiring a captured image from an image acquisition unit based on the adjusted image acquisition parameters.

[0178] The captured image includes an image of a plant to be identified.

[0179] In some embodiments, Fig. 9 As shown, based on the adjusted image acquisition parameters, acquiring the captured image from the image acquisition unit may include:

[0180] Step S131, the image acquisition unit automatically focuses on the determined focus area; and

[0181] Step S132: the image acquisition unit performs photographing to obtain a photographed image.

[0182] That is, after determining the new focus area, the captured image obtained by focusing on the focus area will be captured, so that the final captured image can contain clear plants to be identified, strong characteristic parts of plants to be identified, or distinguishing characteristic parts of plants to be identified, which can help improve the accuracy and efficiency of recognition, so as to improve the recognition effect. It should be noted that focusing on the focus area does not mean not capturing other parts of the preview image, but focusing the shooting on the area in the preview pattern that plays a key role in plant recognition.

[0183] return Figure 2 , the plant identification method may also include:

[0184] Step S140: identifying the plant to be identified based on the captured image.

[0185] Specifically, a plant recognition model based on neural network training may be used to identify plants, thereby helping users determine information such as plant types.

[0186] The present disclosure also proposes a plant identification device, such as Fig.10 As shown, the plant identification device 300 may include a processor 310 and a memory 320. The memory 320 stores instructions. When the instructions are executed by the processor 310, the steps of the plant identification method described above are implemented.

[0187] Among them, the processor 310 can perform various actions and processes according to the instructions stored in the memory 320. Specifically, the processor 310 can be an integrated circuit chip with signal processing capabilities. The above-mentioned processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The various methods, steps and logic block diagrams disclosed in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which can be an X86 architecture or an ARM architecture, etc.

[0188] The memory 320 stores executable instructions, which are executed by the processor 310 to perform the plant identification method described above. The memory 320 can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. The non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous connection dynamic random access memory (SLDRAM) and direct memory bus random access memory (DR RAM). It should be noted that the memory of the method described herein is intended to include but is not limited to these and any other suitable types of memory.

[0189] The present disclosure also provides a plant identification system 400, such as Fig.11 As shown, the plant identification system 400 may include an image acquisition unit 410, an image detection unit 420, at least one of a parameter adjustment unit 430 and a user prompt unit 440, and a plant identification unit 450. The image acquisition unit 410 may be configured to generate a preview image and obtain a captured image, the image detection unit 420 may be configured to detect image acquisition parameters according to the preview image generated by the image acquisition unit 410, the parameter adjustment unit 430 may be configured to automatically adjust the image acquisition parameters according to the detection result, the user prompt unit 440 may be configured to issue a prompt message for instructing the user to adjust the image acquisition parameters according to the detection result, and the plant identification unit 450 may be configured to identify the plant to be identified according to the captured image.

[0190] According to another aspect of the present disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores instructions. When the instructions are executed, the steps in the plant identification method described above can be implemented.

[0191] Similarly, the computer-readable storage medium in the embodiments of the present disclosure may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory. It should be noted that the computer-readable storage medium described herein is intended to include but is not limited to these and any other suitable types of memory.

[0192] The present disclosure also proposes a computer program product, which may include instructions. When the instructions are executed by a processor, the steps of the plant identification method described above can be implemented.

[0193] Instructions may be any set of instructions to be executed directly by one or more processors, such as machine code, or any set of instructions to be executed indirectly, such as a script. The terms "instructions," "application," "process," "step," and "program" herein may be used interchangeably herein. Instructions may be stored in an object code format for direct processing by one or more processors, or as a script or collection of independent source code modules in any other computer language, including those interpreted on demand or compiled in advance. Instructions may include instructions that cause one or more processors, such as to act as the various neural networks herein. The functions, methods, and routines of the instructions are explained in more detail elsewhere herein.

[0194] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0195] In general, various example embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Certain aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device. When various aspects of the embodiments of the present disclosure are illustrated or described as block diagrams, flow charts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general purpose hardware or controllers or other computing devices, or some combination thereof as non-limiting examples.

[0196] The words "front", "rear", "top", "bottom", "above", "below", etc., if present, in the specification and claims are used for descriptive purposes and are not necessarily used to describe invariant relative positions. It should be understood that the words so used are interchangeable under appropriate circumstances, such that the embodiments of the disclosure described herein, for example, are capable of operation in other orientations than those illustrated or otherwise described herein.

[0197] As used herein, the word "exemplary" means "serving as an example, instance, or illustration," rather than as a "model" to be exactly copied. Any implementation described as an example herein is not necessarily to be construed as preferred or advantageous over other implementations. Furthermore, the present disclosure is not limited by any expressed or implied theory given in the above technical field, background technology, summary of the invention, or detailed description.

[0198] As used herein, the term "substantially" is intended to include any minor variations due to design or manufacturing imperfections, device or component tolerances, environmental influences, and / or other factors. The term "substantially" also allows for deviations from a perfect or ideal condition due to parasitic effects, noise, and other practical considerations that may exist in actual implementations.

[0199] In addition, the foregoing description may have referred to elements or nodes or features that are "connected" or "coupled" together. As used herein, unless otherwise expressly stated, "connection" means that one element / node / feature is directly connected (or directly communicates) with another element / node / feature electrically, mechanically, logically or otherwise. Similarly, unless otherwise expressly stated, "coupling" means that one element / node / feature can be mechanically, electrically, logically or otherwise connected to another element / node / feature in a direct or indirect manner to allow interaction, even if the two features may not be directly connected. In other words, "coupling" is intended to include direct and indirect connections of elements or other features, including connections using one or more intermediate elements.

[0200] In addition, the terms "first", "second" and the like may also be used herein for reference purposes only, and thus are not intended to be limiting. For example, the terms "first", "second" and other such numerical terms referring to structures or elements do not imply a sequence or order unless the context clearly indicates otherwise.

[0201] It should also be understood that when the term “include / comprises” is used in this document, it indicates the presence of the specified features, integers, steps, operations, units and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, units and / or components and / or their combinations.

[0202] In this disclosure, the term "provide" is used in a broad sense to cover all ways of obtaining an object, and thus "providing an object" includes but is not limited to "purchasing", "preparing / manufacturing", "arranging / setting up", "installing / assembling", and / or "ordering" an object, etc.

[0203] Although some specific embodiments of the present disclosure have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. The various embodiments disclosed herein may be combined in any manner without departing from the spirit and scope of the present disclosure. It should also be understood by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A plant identification method, It is characterized in that The plant identification method comprises: When the image acquisition unit is in an operating state, detecting image acquisition parameters according to a preview image generated by the image acquisition unit, wherein the image acquisition parameters include at least one of a plant region where a plant is located in the preview image and a part of a plant in the preview image; Automatically adjusting image acquisition parameters according to the detection results and / or issuing prompt information for instructing a user to adjust the image acquisition parameters; Based on the adjusted image acquisition parameters, acquiring a captured image from the image acquisition unit, wherein the captured image includes an image of a plant to be identified; and identifying the plant to be identified according to the captured image, Wherein, automatically adjusting the image acquisition parameters according to the detection result includes: identifying the preliminary classification of the plant to be identified in the preview image, obtaining the strong characteristic part information of the plant to be identified according to the preliminary classification, and determining the area containing the strong characteristic part of the plant to be identified corresponding to the strong characteristic part information as the focus area in the preview image, Based on the adjusted image acquisition parameters, acquiring the captured image from the image acquisition unit includes: the image acquisition unit automatically focusing on the determined focus area, and the image acquisition unit performing shooting to acquire the captured image.

2. The plant identification method according to claim 1, It is characterized in that The image acquisition parameters further include at least one of the following: a shooting distance of the plants in the preview image, a shooting focal length of the plants, and the number of plants.

3. The plant identification method according to claim 1, It is characterized in that Detecting the image acquisition parameters according to the preview image generated by the image acquisition unit includes: A plant region recognition model is used to determine the plant region where the plant in the preview image is located.

4. The plant identification method according to claim 1, It is characterized in that Detecting the image acquisition parameters according to the preview image generated by the image acquisition unit includes: Detecting a plant region where the plant in the preview image is located, and determining an area of ​​the plant region; Calculating the area ratio of the region area to the total area of ​​the preview image; and The shooting distance of the plant is determined according to the area ratio.

5. The plant identification method according to claim 4, It is characterized in that Detecting the plant region where the plant in the preview image is located, and determining the area of ​​the plant region includes: Determine a marked frame circumscribed to the plant, and determine the area of ​​the marked frame as the area of ​​the plant region; or Using a Mask RCNN model to determine a Mask region corresponding to the plant region, and determining the area of ​​the Mask region as the area of ​​the plant region; or A multi-target recognition model is used to respectively determine the plant region where each plant in the preview image is located, and to respectively determine the area of ​​each plant region.

6. The plant identification method according to claim 4, It is characterized in that Automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes: Determining whether the area ratio is within a preset ratio range; and When the area ratio is not within the preset ratio range, at least one of the following operations is performed until the area ratio is within the preset ratio range: Automatically adjust the focus of your shots; and A first prompt message is issued to instruct the user to adjust the shooting distance and / or the shooting focal length.

7. The plant identification method according to claim 4, It is characterized in that Automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes: Determining whether the area ratio is within a preset ratio range; When the area ratio is not within the preset ratio range, automatically adjusting the shooting focal length until the area ratio is within the preset ratio range, or the shooting focal length reaches a maximum focal length or a minimum focal length; and When the shooting focal length reaches the maximum focal length or the minimum focal length, and the area ratio is not within the preset ratio range, a second prompt message is issued to instruct the user to adjust the shooting distance until the area ratio is within the preset ratio range.

8. The plant identification method according to claim 6 or 7, It is characterized in that The plant identification method further comprises: When the area ratio is within the preset ratio range, at least one of the following prompt messages is issued: Third prompt information for indicating to the user that the shooting focal length and shooting distance are appropriate; and The fourth prompt information is used to instruct the user to stop adjusting the shooting focal length and the shooting distance.

9. The plant identification method according to claim 2, It is characterized in that Automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes: When it is detected that the number of plants in the preview image is more than one, fifth prompt information is issued to instruct the user to select at least one of the plants as the plant to be identified; receiving a selection from a user, and determining a plant to be identified in the preview image according to the selection; and An area including at least a portion of the plant to be identified is determined as a focus area in the preview image.

10. The plant identification method according to claim 9, It is characterized in that Automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters also includes: When it is detected that the number of plants in the preview image is more than one, a labeling frame circumscribing each of the plurality of plants is used to label the corresponding plants respectively.

11. The plant identification method according to claim 1, It is characterized in that Detecting the image acquisition parameters according to the preview image generated by the image acquisition unit includes: A plant part recognition model is used to determine the plant part in the preview image.

12. The plant identification method according to claim 1, It is characterized in that Automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes: When it is detected that the plant to be identified in the preview image includes a strong characteristic part, the area including the strong characteristic part is determined as the focus area in the preview image.

13. The plant identification method according to claim 12, It is characterized in that Detecting the image acquisition parameters according to the preview image generated by the image acquisition unit includes: According to a preset priority order, the part of the plant to be identified that has the highest priority and is contained in the preview image is determined as the strong feature part.

14. The plant identification method according to claim 1, It is characterized in that The preliminary classification of the plants to be identified in the preview image includes: A pre-recognition classification model is used to identify the preliminary classification of the plant to be identified in the preview image.

15. The plant identification method according to claim 1, It is characterized in that Acquiring strong characteristic part information of the plant to be identified according to the preliminary classification includes: According to the preliminary classification, the strong characteristic part information of the plant to be identified is retrieved from a first preset database.

16. The plant identification method according to claim 1, It is characterized in that Automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes: Identifying a preliminary classification of the plant to be identified in the preview image; Determining whether the preliminary classification belongs to the easily confused category; When the preliminary classification of the plant to be identified belongs to an easily confused category, determining the distinguishing characteristic part information of the plant to be identified according to the easily confused category; and An area including the distinguishing characteristic part of the plant to be identified and corresponding to the distinguishing characteristic part information is determined as a focus area in the preview image.

17. The plant identification method according to claim 16, It is characterized in that Determining whether the preliminary classification falls into the confusing category includes: According to the preliminary classification, a search is performed from a second preset database to determine whether there is a confusing category corresponding to the preliminary classification.

18. The plant identification method according to claim 16, It is characterized in that Determining the distinguishing characteristic part information of the plant to be identified according to the easily confused category includes: According to the easily confused category, the distinguishing characteristic part information of the plant to be identified is retrieved from a second preset database.

19. The plant identification method according to claim 1, It is characterized in that Automatically adjusting the image acquisition parameters according to the detection result and / or issuing prompt information for instructing the user to adjust the image acquisition parameters includes: When it is detected that the preview image does not include plants, sixth prompt information is issued to instruct the user to adjust the shooting range.

20. The plant identification method according to claim 1, It is characterized in that Identifying the plant to be identified according to the captured image includes: A plant recognition model is used to identify the plant to be identified based on the captured image.

21. A plant identification device, It is characterized in that The plant identification device comprises a processor and a memory, wherein instructions are stored in the memory, and when the instructions are executed by the processor, the steps of the plant identification method according to any one of claims 1 to 20 are implemented.

22. A plant identification system, It is characterized in that The plant identification system comprises: An image acquisition unit, wherein the image acquisition unit is configured to generate a preview image and acquire a captured image; an image detection unit, the image detection unit being configured to detect image acquisition parameters according to the preview image generated by the image acquisition unit, wherein the image acquisition parameters include at least one of a plant region where a plant is located in the preview image and a part of a plant in the preview image; at least one of a parameter adjustment unit and a user prompt unit, wherein the parameter adjustment unit is configured to automatically adjust the image acquisition parameters according to the detection result, and the user prompt unit is configured to issue prompt information for instructing the user to adjust the image acquisition parameters according to the detection result; and a plant identification unit, the plant identification unit being configured to identify the plant to be identified based on the captured image, Wherein, automatically adjusting the image acquisition parameters according to the detection result includes: identifying the preliminary classification of the plant to be identified in the preview image, obtaining the strong characteristic part information of the plant to be identified according to the preliminary classification, and determining the area containing the strong characteristic part of the plant to be identified corresponding to the strong characteristic part information as the focus area in the preview image, Acquiring the captured image includes: automatically focusing on the determined focus area, and photographing to acquire the captured image.

23. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores instructions, and when the instructions are executed, the steps of the plant identification method according to any one of claims 1 to 20 are implemented.

24. A computer program product, It is characterized in that The computer program product comprises instructions, and when the instructions are executed by a processor, the steps of the plant identification method according to any one of claims 1 to 20 are implemented.

Images