Image processing method, storage medium and computer terminal

Abstract

The application discloses an image processing method, a storage medium and a computer terminal. The method comprises the following steps: obtaining an initial spectral image, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; and identifying the target spectral image to determine the category of the target object. The application solves the technical problem of low accuracy of spectral image identification caused by the influence of other spectral signals in the related art.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of image processing, in particular, to an image processing method, a storage medium and a computer terminal. BACKGROUND

[0002] At present, a spectral image is very sensitive to a multi-spectral signal, and the strength of the signal represents the types of different objects in the spectral image. For example, in an agricultural application scenario, a spectral image of a crop contains multiple crop types, but due to the influence of clouds, light, and the like in later processing, the same crop has different spectral signal values, thereby causing the crop recognition and classification accuracy to be reduced.

[0003] In view of the above problems, an effective solution has not been proposed yet. SUMMARY

[0004] Embodiments of the present application provide an image processing method, a storage medium and a computer terminal to at least solve the technical problem of low accuracy of spectral image recognition due to the influence of other spectral signals in the related art.

[0005] According to an aspect of embodiments of the present application, an image processing method is provided, including: obtaining an initial spectral image, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; and identifying the target spectral image to determine a category of the target object.

[0006] According to another aspect of embodiments of the present application, an image processing method is also provided, including: obtaining an initial spectral image, wherein the initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; and identifying the target spectral image to determine a category of the target crop.

[0007] According to another aspect of embodiments of the present application, an image processing method is also provided, including: obtaining an initial spectral image, wherein the initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; and identifying the target spectral image to determine a yield of the target crop.

[0008] According to another aspect of the embodiments of the present application, an image processing method is also provided. The method comprises: a cloud server receiving an initial spectral image uploaded by a client, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; the cloud server superimposes spectral noise on the initial spectral image to obtain a target spectral image; and the cloud server identifies the target spectral image to determine the category of the target object.

[0009] According to an aspect of the embodiments of the present application, an image processing apparatus is provided. The apparatus comprises: an acquisition module configured to acquire an initial spectral image, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; a superimposition module configured to superimpose spectral noise on the initial spectral image to obtain a target spectral image; and an identification module configured to identify the target spectral image to determine the category of the target object.

[0010] According to another aspect of the embodiments of the present application, an image processing apparatus is also provided. The apparatus comprises: an acquisition module configured to acquire an initial spectral image, wherein the initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; a superimposition module configured to superimpose spectral noise on the initial spectral image to obtain a target spectral image; and an identification module configured to identify the target spectral image to determine the category of the target crop.

[0011] According to another aspect of the embodiments of the present application, an image processing apparatus is also provided. The apparatus comprises: an acquisition module configured to acquire an initial spectral image, wherein the initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; a superimposition module configured to superimpose spectral noise on the initial spectral image to obtain a target spectral image; and an identification module configured to identify the target spectral image to determine the yield of the target crop.

[0012] According to another aspect of the embodiments of the present application, an image processing apparatus is also provided. The apparatus comprises: a receiving module configured to receive, by a cloud server, an initial spectral image uploaded by a client, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; a superimposition module configured to superimpose, by the cloud server, spectral noise on the initial spectral image to obtain a target spectral image; and an identification module configured to identify, by the cloud server, the target spectral image to determine the category of the target object.

[0013] According to another aspect of the embodiments of the present application, a computer readable storage medium is also provided. The computer readable storage medium comprises a stored program, wherein the program, when executed, controls a device in which the computer readable storage medium is located to perform the image processing method described above.

[0014] According to another aspect of the embodiments of the present application, a computer terminal is further provided, comprising a memory and a processor, the processor being configured to execute a program stored in the memory, wherein the program performs the image processing method described above when executed.

[0015] In the embodiments of the present application, an initial spectral image is acquired, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; the spectral noise is superimposed on the initial spectral image to obtain a target spectral image; and the target spectral image is identified to determine the category of the target object, thereby improving the identification accuracy of the object in the spectral image. It is easy to note that the spectral noise can be used to improve the spectral adaptation range of the same type of object in the spectral image, and by increasing the difference between different spectrums, the influence of other factors can be eliminated, thereby improving the identification accuracy of the object in the spectral image, and further solving the technical problem of low identification accuracy of the spectral image in the related art due to the influence of other spectral signals. BRIEF DESCRIPTION OF DRAWINGS

[0016] The accompanying drawings, which are included to provide a further understanding of the present application, form a part of the present application and illustrate the illustrative embodiments of the present application and together with the description serve to explain the present application. In the drawings:

[0017] Figure 1 is a hardware structure block diagram of a computer terminal (or mobile device) for implementing an image processing method according to the prior art;

[0018] Figure 2 is a flowchart of an image processing method according to Embodiment 1 of the present application;

[0019] Figure 3 is a flowchart of another image processing method according to Embodiment 1 of the present application;

[0020] Figure 4a is a schematic diagram of a hyperspectral image of a crop according to Embodiment 1 of the present application;

[0021] Figure 4b is a structural schematic diagram of an image recognition process according to Embodiment 1 of the present application;

[0022] Figure 5 is a schematic diagram of an image processing process according to Embodiment 1 of the present application;

[0023] Figure 6 is a flowchart of an image processing method according to Embodiment 2 of the present application;

[0024] Figure 7is a flow chart of an image processing method according to an embodiment of the present application;

[0025] Figure 8 is a schematic diagram of an image processing apparatus according to an embodiment of the present application;

[0026] Figure 9 is a schematic diagram of an image processing apparatus according to an embodiment of the present application;

[0027] Figure 10 is a schematic diagram of an image processing apparatus according to an embodiment of the present application;

[0028] Figure 11 is a flow chart of an image processing method according to an embodiment of the present application;

[0029] Figure 12 is a schematic diagram of an image processing apparatus according to an embodiment of the present application;

[0030] Figure 13 is a structural block diagram of a computer terminal according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to make the personnel in the art better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0032] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0033] First, some of the nouns or terms appearing in the description of the embodiments of the present application are applicable to the following explanations:

[0034] NDVI (Normalized difference vegetation index): Normalized Difference Vegetation Index

[0035] NIR (Near infrared band): Near Infrared

[0036] MSB (Multi spectral band): Multi spectral band.

[0037] With the fierce global trade competition, the prices of agricultural commodities such as agriculture fluctuate sharply, and agricultural security and grain yield statistics provide key data reference for the formulation of national agricultural strategy and agricultural planning. Crop recognition is the core technology of agricultural asset inventory, crop trend detection, and agricultural disaster insurance loss assessment. The result of crop recognition directly affects the accuracy of subsequent evaluation, so improving the accuracy of crop recognition plays a key role in promoting agricultural statistics planning.

[0038] At present, remote sensing crop distribution recognition is very sensitive to multi-spectral signals, and the strength of the signal represents different crop types. Due to clouds, light, image post-processing and other images, the same crop has different spectral signal values, resulting in reduced accuracy of crop recognition and classification. The present application provides a processing method based on relative spectral normalization, which makes the crop invariant under the conditions of cloud, light anomaly, different image post-processing, thereby improving the accuracy of crop recognition of wheat, corn, soybean and other crops.

[0039] To solve the above problems, the present application provides an image processing method, which can enhance the generalization performance of the deep network through spectral gain, and improve the recognition accuracy of the model; based on the relative normalization of the spectrum, the relative value is used as the network input, which can solve the influence caused by thin fog, shadow, and post-image truncation processing, thereby improving the recognition accuracy of the image.

[0040] Embodiment 1

[0041] According to the embodiments of the present application, an embodiment of an image processing method is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0042] The method embodiments provided by the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing an image processing method is shown. As shown in FIG. 1, the computer terminal includes a processor 101, a memory 102, a storage 103, a communication interface 104, a display 105, and the like. Figure 1As shown, the computer terminal 10 (or mobile device 10) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0043] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuitry are generally referred to herein as "data processing circuitry". This data processing circuitry may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuitry may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). This data processing circuitry serves as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0044] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the image processing method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the above-mentioned image processing method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0045] The transmission device 106 is configured to receive or send data via a network. The network can include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network interface controller (NIC) that can connect to other network devices through a base station to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module that is configured to communicate with the Internet wirelessly.

[0046] The display can be a liquid crystal display (LCD) that is touch screen, for example, which can enable a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0047] It is noted that in some alternative embodiments, the above-mentioned Figure 1 The computer device (or mobile device) can include hardware elements (including circuitry), software elements (including computer code stored on a computer readable medium), or a combination of both hardware and software elements. It should be noted that in some embodiments, the functions of the computer device (or mobile device) can be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output. Figure 1 is merely one example of a particular implementation and is intended to illustrate the types of components that can be present in the above-mentioned computer device (or mobile device).

[0048] In the above-mentioned operating environment, the present application provides an image processing method as shown in the accompanying drawings. Figure 2 Figure 2 is a flowchart of an image processing method according to an embodiment of the present application. As shown in the accompanying drawings, the method includes the following steps: Figure 2

[0049] In step S202, an initial spectral image is obtained.

[0050] The initial spectral image contains a target object, and is composed of image information and first spectral information of a plurality of spectral channels.

[0051] The initial spectral image can be a spectral image in an agricultural and forestry application scenario, and the target object contained therein can be crops. The initial spectral image can also be a spectral image in a city planning scenario, and the target object contained therein can be a building. The initial spectral image can also be a spectral image in a transportation scenario, and the target object contained therein can be a vehicle. The initial spectral image can also be a spectral image in a natural resource scenario, and the target object contained therein can be a land plot or a water body. The initial spectral image can be a hyperspectral image or a multispectral image. The present application is described by taking an agricultural scenario as an example.

[0052] ​​In an optional embodiment, the initial spectral image can be taken by a satellite or a drone and transmitted to the server through the network, processed by the server, and displayed to the user, and the initial spectral image can be displayed in the image collection area. In another optional embodiment, the initial spectral image can be taken by a satellite or a drone and uploaded to the server by the user, processed by the server, and the user can upload the initial spectral image to the server by clicking the "upload image" button in the interactive interface or directly dragging the initial spectral image into the dashed box. Moreover, the uploaded initial spectral image can be displayed in the image collection area. The server here can be a server deployed locally or a server deployed in the cloud.

[0053] Step S204: superimposing the spectral noise on the initial spectral image to obtain a target spectral image.

[0054] The spectral noise described above can be randomly selected noise.

[0055] In an optional embodiment, the initial spectral image may, due to natural lighting conditions, ground topography, and other reasons, introduce various noises during imaging, which can cause the target objects contained in multiple initial spectral images to have different spectral signal values, thereby affecting the recognition accuracy of the target objects. In order to eliminate the influence of environmental factors on the target objects, the spectral adaptation range of the target objects can be increased by means of spectral noise in order to increase the difference between different spectra, thereby improving the recognition accuracy of the target spectral image.

[0056] In another optional embodiment, the spectral noise can be superimposed on the image information in the initial spectral image in order to increase the spectral adaptation range of the target objects on the overall level of the initial spectral image and increase the difference between different spectra of the target objects.

[0057] In another optional embodiment, the spectral noise can be superimposed on the first spectral information of each spectral channel in the initial spectral image. Since the spectral noise is randomly superimposed, the spectral noise superimposed on each spectral channel can be different, which can further increase the spectral adaptation range of the target objects in the initial spectral image, thereby improving the recognition accuracy of the target objects in the target spectral image.

[0058] In yet another optional embodiment, in order to further improve the recognition accuracy of the target objects, the image information or the first spectral information of multiple spectral channels can be enhanced, and the enhanced image information or the first spectral information of multiple spectral channels can be used to superimpose the spectral noise, thereby improving the recognition accuracy of the target objects in the target spectral image.

[0059] In step S206, the target spectral image is recognized to determine the category of the target object.

[0060] In an optional embodiment, the second spectral information of each spectral channel in the target spectral image can be acquired, and the second spectral information of each spectral image is normalized to obtain a normalized image. Since the target spectral image is obtained by superimposing spectral noise, the characteristics of the second spectral information of each spectral channel are more obvious. By normalizing the second spectral information, information belonging to the same category can be unified, thereby reducing the influence caused by environmental factors.

[0061] It should be noted that the target spectral image can not only be recognized to determine the category of the target object, but also the number of target objects can be recognized. For example, for the spectral image of the agricultural and forestry application scene, not only the category of the crop can be recognized, but also the crop yield can be estimated.

[0062] In another optional embodiment, after determining the category of the target object, the target object can be labeled according to different colors to distinguish the multiple target objects in the target spectral image.

[0063] In another optional embodiment, after recognizing the target spectral image, the server can directly display the recognition result for the user to view. The category of the target object displayed in the recognition result can be displayed in the target spectral image in the result feedback area. In another optional embodiment, after recognizing the target spectral image, the server can feed back the recognition result to the client of the user through the network, and the client can display the recognition result for the user to view. Specifically, the client can display the category of the target object on the target spectral image. If the user thinks that the category of the target object in the detection result is incorrect, the user can modify the category of the target object, re-input the category of the target object, and upload it to the server. Thus, the server can re-train the model used for target recognition according to the feedback of the user, so as to improve the performance of the server.

[0064] In another optional embodiment, the target spectral image can be recognized by an image segmentation network to determine the category of the target object.

[0065] In the agricultural and forestry application scenario, the initial spectral image can be a crop image, the target object contained in the crop image can be soybean, corn, etc., and the spectral noise can be cloud, light, and other spectra affecting recognition accuracy. The crop image can be obtained first, and then the image information of the crop image or the first spectral information of the plurality of spectral channels is noise enhanced, and the enhanced spectral image is superimposed with the initial spectral image, the spectral adaptation range of the same crop is improved, the difference between different spectra in the target spectral image is increased, after obtaining the target spectral image, the second spectral information of the plurality of spectral channels in the target spectral image is normalized to obtain the relative relationship of different channels, to eliminate the interference of cloud, light, and other images, thereby improving the recognition accuracy of the crop in the crop image. After determining the category of the target object in the crop image, different colors can be marked for each category of crop to distinguish different types of crops.

[0066] In the city planning scenario, the initial spectral image can be a building image, the target object contained in the building image can be a residence, a commercial building, etc., and the spectral noise can be cloud, light, and other spectra affecting recognition accuracy. The building image can be obtained first, and then the image information of the building image or the first spectral information of the plurality of spectral channels is noise enhanced, and the enhanced spectral image is superimposed with the initial spectral image, the spectral adaptation range of the same building is improved, the difference between different spectra in the target spectral image is increased, after obtaining the target spectral image, the second spectral information of the plurality of spectral channels in the target spectral image is normalized to obtain the relative relationship of different channels, to eliminate the interference of cloud, light, and other images, thereby improving the recognition accuracy of the building in the building image. After determining the category of the target object in the building image, different colors can be marked for each category of building to distinguish different types of buildings.

[0067] In a natural resource scene, the initial spectral image can be a land image, the target object contained in the land image can be farmland, grassland, forest land, and the like, and the spectral noise can be cloud, light, and the like, which affect the recognition accuracy. The land image can be acquired first, then the image information of the land image or the first spectral information of the plurality of spectral channels is subjected to noise enhancement, and the enhanced spectral image is superimposed with the initial spectral image, the spectral adaptation range of the same crop is improved, the difference between different spectrums in the target spectral image is increased, after the target spectral image is obtained, the second spectral information of the plurality of spectral channels in the target spectral image is subjected to normalization processing, the relative relationship of different channels is obtained, the interference of cloud, light, and the like is eliminated, and thus the recognition accuracy of the land in the land image is improved. After the category of the target object in the land image is determined, the land of each category can be marked with different colors to distinguish different types of land.

[0068] Through the above steps, the initial spectral image is acquired first, the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; the spectral noise is superimposed with the initial spectral image to obtain a target spectral image; and the target object is identified to determine the category of the target object, thereby improving the recognition accuracy of the object in the spectral image. It is easy to note that the spectral adaptation range of the same type of object in the spectral image can be improved by using the spectral noise, the influence of other factors can be eliminated by increasing the difference between different spectrums, thereby improving the recognition accuracy of the object in the spectral image, and thus the technical problem of low recognition accuracy of the spectral image caused by other spectral signals in the related art is solved.

[0069] In the above embodiments of the present application, superimposing the spectral noise with the initial spectral image to obtain the target spectral image includes at least one of the following: superimposing the image information with the spectral noise to obtain the target spectral image; and superimposing the first spectral information of the plurality of spectral channels with the spectral noise to obtain the target spectral image.

[0070] In an optional embodiment, the initial spectral image can be subjected to overall spectral gain. Specifically, the image information of the initial spectral image can be multiplied by the noise intensity of the spectral noise, and then superimposed with the spectral noise to realize overall spectral gain of the initial spectral image, thereby improving the adaptation range of the same type of object in the initial spectral image, increasing the difference between different spectrums, and reducing the influence of noise on the target object in the target spectral image.

[0071] In another optional embodiment, independent spectral gain can be performed on the initial spectral image, specifically, the first spectral information of the plurality of spectral channels in the initial spectral image can be multiplied by the spectral noise and the noise intensity of the spectral noise, and then superimposed with the spectral noise to obtain the target spectral image, so as to perform gain on each spectral channel in the initial spectral image, thereby improving the adaptation range of the same type of object in the initial spectral image, so as to increase the difference between different spectrums, thereby reducing the noise impact on the target object in the target spectral image.

[0072] By performing spectral gain on the initial spectral image in the above two manners, the generalization performance of the recognition network can be enhanced, thereby improving the recognition accuracy of the target object.

[0073] In the above embodiments of the present application, superimposing the image information and the spectral noise to obtain the target spectral image includes: obtaining the product of the image information and the noise intensity of the spectral noise to obtain a first product; and obtaining the sum of the first product and the spectral noise to obtain the target spectral image.

[0074] In an optional embodiment, the initial spectral image can be subjected to spectral gain by the following formula to obtain the target spectral image.

[0075] Image2 = Image1 * RandomScale + RandomNoise

[0076] wherein, Image1 is the image information of the initial spectral image, Image2 is the image information of the target spectral image, RandomScale is the noise intensity of the spectral noise, and RandomNoise is the spectral noise.

[0077] In the above embodiments of the present application, superimposing the first spectral information of the plurality of spectral channels and the spectral noise to obtain the target spectral image includes: obtaining the product of the first spectral information of each spectral channel and the noise intensity of the spectral noise to obtain a first product corresponding to each spectral channel; and obtaining the sum of the first product corresponding to each spectral channel and the spectral noise to obtain the second spectral information of each spectral channel contained in the target spectral image.

[0078] In an optional embodiment, the initial spectral image can be subjected to spectral gain by the following formula to obtain the target spectral image.

[0079] Bi2 = Bi1 * RandomScale + RandomNoise

[0080] Bi1 is first spectral information of a plurality of spectral channels in the initial spectral image, Bi2 is second spectral information of the plurality of spectral channels in the target spectral image, RandomScale is a noise intensity of the spectral noise, and RandomNoise is the spectral noise.

[0081] In the above embodiments of the present application, after the spectral noise is superimposed on the initial spectral image to obtain the target spectral image, the method further includes: performing normalization processing on the second spectral information of each spectral channel contained in the target spectral image to obtain a normalized image; and performing identification on the normalized image to determine the category of the target object.

[0082] In an optional embodiment, the second spectral information of each spectral channel contained in the target spectral image can be normalized, specifically, the normalization can be performed according to a wave band relative value; the normalization can also be performed according to a wave band interval value, and the spectral can be divided into a plurality of intervals, and the plurality of channels can be normalized according to the intervals. Since the pixel value can be affected by environmental factors, for example, affected by light intensity, therefore, the relative value or the relative value of the interval can be used to eliminate the influence of the environmental factors, thereby improving the identification accuracy of the category of the target object.

[0083] In the above embodiments of the present application, the normalization processing on the second spectral information of each spectral channel contained in the target spectral image to obtain the normalized image includes at least one of the following: performing normalization processing on the second spectral information of adjacent spectral channels to obtain the normalized image; and performing normalization processing on the second spectral information of the spectral channels belonging to the same spectral channel interval to obtain the normalized image.

[0084] In an optional embodiment, the normalization processing can be performed on the second spectral information of adjacent spectral channels to obtain the relative relationship between different channels, thereby eliminating the influence of environmental factors on the target spectral image.

[0085] The above spectral channel interval can be three interval wave bands of visible light (bgr), infrared light (nir, nir1, nir2), and yellow-violet light.

[0086] In another optional embodiment, the normalization can be performed on the second spectral information of the spectral channels belonging to the same spectral channel interval according to the interval value of the wave band interval, so as to normalize different channels into corresponding intervals, so as to eliminate the influence of other environmental factors.

[0087] In the above embodiments of the present application, the second spectral information of the adjacent spectral channels is normalized to obtain the normalized image, including: obtaining the difference between the second spectral information of the adjacent spectral channels to obtain a spectral difference value; obtaining the sum of the second spectral information of the adjacent spectral channels to obtain a spectral sum value; obtaining the ratio of the spectral difference value and the spectral sum value to obtain the normalized image.

[0088] In an alternative embodiment, the second spectral information of the adjacent spectral channels can be normalized by (Bi-Bj) / (Bi+Bj) to obtain the normalized image. Wherein, Bi and Bj are the second spectral information of the adjacent spectral channels.

[0089] In the above embodiments of the present application, the second spectral information of the spectral channels belonging to the same spectral channel interval is normalized to obtain the normalized image, including: determining the target spectral channel interval to which each spectral channel belongs; obtaining the sum of at least one spectral channel corresponding to the target spectral channel interval to obtain interval spectral information; obtaining the ratio of the second spectral information of each spectral channel and the interval spectral information to obtain the normalized image.

[0090] In an alternative embodiment, the second spectral information of the spectral channels belonging to the same spectral channel interval can be normalized by Br / (Br+Bg+Br) to obtain the normalized image. Br is the second spectral information of any one spectral channel, and Br+Bg+Br is the interval spectral information.

[0091] In the above embodiments of the present application, after the spectral noise is superimposed on the initial spectral image to obtain the target spectral image, the method further includes: outputting the initial spectral image and the target spectral image; receiving feedback information, wherein the feedback information is used to represent whether the spectral noise is superimposed on the initial spectral image; in the case that the feedback information is to superimpose the spectral noise on the initial spectral image, identifying the target spectral image to determine the category of the target object.

[0092] In an optional embodiment, after superimposing the spectral noise on the initial spectral image, the initial spectral image and the target spectral image can be displayed for the user to view, or sent to the client of the user for display for the user to view. The user can determine whether to superimpose the spectral noise on the initial spectral image according to the processing effect of the target spectral image, and return corresponding feedback information, i.e., whether the target spectral image needs to be identified to determine the type of the target object. If the user determines that superposition is needed, feedback information of superimposing the spectral noise on the initial spectral image can be returned, at which time, the target spectral image can be identified to determine the category of the target object. If the user determines that superposition is not needed, feedback information of not superimposing the spectral noise on the initial spectral image can be returned, at which time, the initial spectral image can be directly identified to determine the category of the target object.

[0093] In the above embodiments of the present application, after identifying the target spectral image to determine the category of the target object, the method further includes: determining a target display mode corresponding to the target object based on the category of the target object; and marking the target object in the target spectral image according to the target display mode.

[0094] The target display mode described above can be one or more of a display color, a display mark (e.g., a bounding box), a related text displaying the category of the target object, etc., which can be set according to actual needs.

[0095] In an optional embodiment, in order to facilitate the user to intuitively and accurately distinguish different categories of target objects, different display colors can be set in advance for different categories, so that after determining the category of the target object, the target object can be marked in the target spectral image according to the corresponding display color. For example, taking the agricultural and forestry application scenario as an example, the target objects contained in the crop image can be soybeans, corn, etc., and the corresponding colors are red and yellow. Therefore, after determining that the target type of the target object is corn, the area where the target object is located in the crop image can be filled with yellow.

[0096] In another optional embodiment, in order to facilitate the user to intuitively and accurately distinguish different categories of target objects, different colors, different line widths, and different line types of bounding boxes can be set in advance for different categories, so that after determining the category of the target object, the target object can be marked in the target spectral image using the corresponding bounding box. For example, still taking the agricultural and forestry application scenario as an example, the target objects contained in the crop image can be soybeans, corn, etc., and the corresponding bounding box colors are red and yellow. Therefore, after determining that the target type of the target object is corn, a yellow square can be used to frame the area where the target object is located in the crop image.

[0097] In another optional embodiment, in order to help users intuitively and accurately distinguish different categories of target objects, the corresponding text of the category can be directly displayed in the target spectral image. For example, taking the agricultural and forestry application scenario as an example, the target objects contained in the crop image can be soybeans, corn, etc. Therefore, after determining that the target type of the target object is corn, the text "corn" can be marked next to the area where the target object is located in the crop image.

[0098] The following is combined Figures 3 to 5 A preferred embodiment of this application will be described in detail. This method can be executed by a computer terminal or a server. For example... Figure 3 As shown, the method includes the following steps:

[0099] Step S301: Obtain hyperspectral images of crops;

[0100] Optionally, the hyperspectral image described above is the initial spectral image. For example... Figure 4a The image shown is a hyperspectral image of a crop.

[0101] Step S302: Perform overall spectral gain or independent spectral gain on the hyperspectral image of the crop to obtain the target spectral image;

[0102] Step S303 involves normalizing the spectral information of multiple spectral channels in the target spectral image by band relative value or band interval value to obtain a normalized image.

[0103] Step S304: Recognize the normalized image to determine the category of crops.

[0104] like Figure 4b The diagram shows the crop identification results, which include soybeans, corn, and rice.

[0105] like Figure 5 The diagram shows the structure of the image recognition process. A certain type of hyperspectral remote sensing data is selected as input. First, spectral data enhancement is performed, then spectral normalization is performed to obtain the input data for the deep network, and finally the data is sent to the image segmentation network to obtain the final crop classification result.

[0106] In view of the poor crop recognition effect of hyperspectral images, the model expansion generalization and image quality interference problems such as clouds and fog in crop recognition can be solved by two methods of spectral data enhancement and spectral normalization. First, the spectral data is enhanced, the spectral noise is randomly increased, the spectral adaptation range of the same crop is improved, and the difference between different spectrums is increased. Second, the spectral normalization is performed, and the relative relationship of different channels is obtained in a manner similar to the NDVI (Normalized Difference Vegetation Index), so as to eliminate image interference such as shadows, clouds and fog, and improve the accuracy of crop recognition.

[0107] It should be noted that, for the foregoing method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the action sequence described, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily necessary for the present application.

[0108] From the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes a plurality of instructions for making a terminal device (which can be a mobile phone, computer, server, or network device, etc.) execute the method of each embodiment of the present application.

[0109] Embodiment 2

[0110] According to the embodiments of the present application, an image processing method embodiment is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0111] Figure 6 is a flowchart of an image processing method according to Embodiment 2 of the present application. As shown in Figure 6 , the method can include the following steps:

[0112] Step S602, acquiring an initial spectral image.

[0113] The initial spectral image contains the target crop and is composed of image information and first spectral information of a plurality of spectral channels.

[0114] In step S604, the spectral noise is superimposed on the initial spectral image to obtain a target spectral image.

[0115] In step S606, the target spectral image is identified to determine the category of the target crop.

[0116] In the above embodiments of the present application, superimposing the spectral noise on the initial spectral image to obtain the target spectral image includes at least one of the following: superimposing the image information and the spectral noise to obtain the target spectral image; and superimposing the first spectral information of the plurality of spectral channels and the spectral noise to obtain the target spectral image.

[0117] In the above embodiments of the present application, superimposing the image information and the spectral noise to obtain the target spectral image includes: obtaining a product of the image information and a noise intensity of the spectral noise to obtain a first product; and obtaining a sum of the first product and the spectral noise to obtain the target spectral image.

[0118] In the above embodiments of the present application, superimposing the first spectral information of the plurality of spectral channels and the spectral noise to obtain the target spectral image includes: obtaining a product of the first spectral information of each spectral channel and a noise intensity of the spectral noise to obtain a first product corresponding to each spectral channel; and obtaining a sum of the first product corresponding to each spectral channel and the spectral noise to obtain second spectral information of each spectral channel contained in the target spectral image.

[0119] In the above embodiments of the present application, after superimposing the spectral noise on the initial spectral image to obtain the target spectral image, the method further includes: performing normalization processing on the second spectral information of each spectral channel contained in the target spectral image to obtain a normalized image; and identifying the normalized image to determine the category of the target crop.

[0120] In the above embodiments of the present application, performing normalization processing on the second spectral information of each spectral channel contained in the target spectral image to obtain the normalized image includes at least one of the following: performing normalization processing on the second spectral information of adjacent spectral channels to obtain the normalized image; and performing normalization processing on the second spectral information of spectral channels belonging to the same spectral channel interval to obtain the normalized image.

[0121] In the above embodiments of the present application, performing normalization processing on the second spectral information of adjacent spectral channels to obtain the normalized image includes: obtaining a difference between the second spectral information of the adjacent spectral channels to obtain a spectral difference value; obtaining a sum of the second spectral information of the adjacent spectral channels to obtain a spectral sum value; and obtaining a ratio of the spectral difference value and the spectral sum value to obtain the normalized image.

[0122] In the above embodiment of the present application, the second spectral information of the spectral channels belonging to the same spectral channel interval is normalized to obtain a normalized image, including: determining the target spectral channel interval to which each spectral channel belongs; obtaining the sum of at least one spectral channel corresponding to the target spectral channel interval to obtain interval spectral information; obtaining the ratio of the second spectral information of each spectral channel to the interval spectral information to obtain a normalized image.

[0123] In the above embodiment of the present application, after superimposing the spectral noise on the initial spectral image to obtain the target spectral image, the method further includes: outputting the initial spectral image and the target spectral image; receiving feedback information, wherein the feedback information is used to represent whether to superimpose the spectral noise on the initial spectral image; in the case that the feedback information is to superimpose the spectral noise on the initial spectral image, identifying the target spectral image to determine the category of the target crop.

[0124] In the above embodiment of the present application, after identifying the target spectral image to determine the category of the target crop, the method further includes: determining the target display mode corresponding to the target crop based on the category of the target crop; and marking the target crop in the target spectral image according to the target display mode.

[0125] It should be noted that the preferred embodiments involved in the above embodiments of the present application have the same application scenarios and implementation processes as the schemes provided in Embodiment 1, but are not limited to the schemes provided in Embodiment 1.

[0126] Embodiment 3

[0127] According to the embodiments of the present application, an image processing method embodiment is also provided. It should be noted that the steps shown in the flowchart can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from here.

[0128] Figure 7 is a flowchart of an image processing method according to Embodiment 3 of the present application. As shown in Figure 7 the method can include the following steps:

[0129] Step S702, the cloud server receives the initial spectral image uploaded by the client.

[0130] The initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels.

[0131] Step S704, the cloud server superimposes the spectral noise on the initial spectral image to obtain a target spectral image.

[0132] In step S706, the cloud server identifies the target spectral image to determine the category of the target object.

[0133] In the above embodiments, the cloud server superimposes the spectral noise and the initial spectral image to obtain the target spectral image, including at least one of the following: the cloud server superimposes the image information and the spectral noise to obtain the target spectral image; and the cloud server superimposes the first spectral information of the plurality of spectral channels and the spectral noise to obtain the target spectral image.

[0134] In the above embodiments, the cloud server superimposes the image information and the spectral noise to obtain the target spectral image, including: the cloud server obtains a product of the image information and the noise intensity of the spectral noise to obtain a first product; and the cloud server obtains a sum of the first product and the spectral noise to obtain the target spectral image.

[0135] In the above embodiments, the cloud server superimposes the first spectral information of the plurality of spectral channels and the spectral noise to obtain the target spectral image, including: the cloud server obtains a product of the first spectral information of each spectral channel and the noise intensity of the spectral noise to obtain a first product corresponding to each spectral channel; and the cloud server obtains a sum of the first product corresponding to each spectral channel and the spectral noise to obtain the second spectral information of each spectral channel included in the target spectral image.

[0136] In the above embodiments, after the cloud server superimposes the spectral noise and the initial spectral image to obtain the target spectral image, the method further includes: the cloud server performs normalization processing on the second spectral information of each spectral channel included in the target spectral image to obtain a normalized image; and the cloud server identifies the normalized image to determine the category of the target object.

[0137] In the above embodiments, the cloud server performs normalization processing on the second spectral information of each spectral channel included in the target spectral image to obtain the normalized image, including at least one of the following: the cloud server performs normalization processing on the second spectral information of adjacent spectral channels to obtain the normalized image; and the cloud server performs normalization processing on the second spectral information of spectral channels belonging to the same spectral channel interval to obtain the normalized image.

[0138] In the above embodiments, the cloud server performs normalization processing on the second spectral information of adjacent spectral channels to obtain the normalized image, including: the cloud server obtains a difference between the second spectral information of adjacent spectral channels to obtain a spectral difference value; the cloud server obtains a sum of the second spectral information of adjacent spectral channels to obtain a spectral sum value; and the cloud server obtains a ratio of the spectral difference value and the spectral sum value to obtain the normalized image.

[0139] In the above embodiment of the present application, the cloud server performs normalization processing on the second spectral information of the spectral channels belonging to the same spectral channel interval to obtain a normalized image, including: the cloud server determines a target spectral channel interval to which each spectral channel belongs; the cloud server obtains the sum of at least one spectral channel corresponding to the target spectral channel interval to obtain interval spectral information; and the cloud server obtains the ratio of the second spectral information of each spectral channel to the interval spectral information to obtain a normalized image.

[0140] In the above embodiment of the present application, after superimposing the spectral noise on the initial spectral image to obtain the target spectral image, the method further includes: outputting the initial spectral image and the target spectral image; receiving feedback information, wherein the feedback information is used to represent whether to superimpose the spectral noise on the initial spectral image; and in the case that the feedback information is to superimpose the spectral noise on the initial spectral image, identifying the target spectral image to determine the category of the target object.

[0141] In the above embodiment of the present application, after identifying the target spectral image to determine the category of the target object, the method further includes: determining a target display mode corresponding to the target object based on the category of the target object; and marking the target object in the target spectral image according to the target display mode.

[0142] In the above embodiment of the present application, the cloud server outputs the target detection result, including: the cloud server obtains a preset display mode corresponding to the category of the target object; and the cloud server outputs the target detection result according to the preset display mode.

[0143] Embodiment 4

[0144] According to the embodiments of the present application, an image processing device for implementing the above image processing method is also provided, as shown in Figure 8 The device 800 includes an acquisition module 802, a superposition module 804, and an identification module 806.

[0145] The acquisition module 802 is configured to acquire an initial spectral image, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; the superposition module 804 is configured to superimpose spectral noise on the initial spectral image to obtain a target spectral image; and the identification module 806 is configured to identify the target spectral image to determine the category of the target object.

[0146] It should be noted that the above acquisition module 802, superposition module 804, identification module 806 correspond to steps S202 to S206 in Embodiment 1, the three modules have the same instances and application scenarios as the corresponding steps, but are not limited to the above disclosed content of Embodiment 1. It should be noted that the above modules can run in the computer terminal 10 provided in Embodiment 1 as part of the device.

[0147] In the above embodiments of the present application, the superposition module comprises at least one of: a first superposition unit, a second superposition unit.

[0148] The first superposition unit is configured to superimpose the image information and the spectral noise to obtain the target spectral image; and the second superposition unit is configured to superimpose the first spectral information of the plurality of spectral channels and the spectral noise to obtain the target spectral image.

[0149] In the above embodiments of the present application, the first superposition unit comprises: a first acquisition subunit, a second acquisition subunit.

[0150] The first acquisition subunit is configured to obtain a product of the image information and a noise intensity of the spectral noise to obtain a first product; and the second acquisition subunit is configured to obtain a sum of the first product and the spectral noise to obtain the target spectral image.

[0151] In the above embodiments of the present application, the second superposition unit comprises: a third acquisition subunit, a fourth acquisition subunit.

[0152] The third acquisition subunit is configured to obtain a product of the first spectral information of each spectral channel and a noise intensity of the spectral noise to obtain a first product corresponding to each spectral channel; and the fourth acquisition subunit is configured to obtain a sum of the first product corresponding to each spectral channel and the spectral noise to obtain the second spectral information of each spectral channel contained in the target spectral image.

[0153] In the above embodiments of the present application, the device further comprises: a normalization processing module configured to perform normalization processing on the second spectral information of each spectral channel contained in the target spectral image to obtain a normalized image; and the identification module is further configured to identify the normalized image to determine the category of the target object.

[0154] In the above embodiments of the present application, the normalization processing module comprises at least one of: a first normalization processing unit, a second normalization processing unit.

[0155] The first normalization processing unit is configured to perform normalization processing on the second spectral information of adjacent spectral channels to obtain the normalized image; and the second normalization processing unit is configured to perform normalization processing on the second spectral information of the spectral channels belonging to the same spectral channel interval to obtain the normalized image.

[0156] In the above embodiment of the present application, the second normalization processing unit comprises: a first obtaining subunit, a second obtaining subunit, and a third obtaining subunit.

[0157] The first obtaining subunit is configured to obtain a difference between the second spectral information of adjacent spectral channels to obtain a spectral difference value; the second obtaining subunit is configured to obtain a sum of the second spectral information of adjacent spectral channels to obtain a spectral sum value; and the third obtaining subunit is configured to obtain a ratio of the spectral difference value and the spectral sum value to obtain the normalized image.

[0158] In the above embodiment of the present application, the second normalization processing unit comprises: a determining subunit, a fourth obtaining subunit, and a fifth obtaining subunit.

[0159] The determining subunit is configured to determine a target spectral channel interval to which each spectral channel belongs; the fourth obtaining subunit is configured to obtain a sum of at least one spectral channel corresponding to the target spectral channel interval to obtain interval spectral information; and the fifth obtaining subunit is configured to obtain a ratio of the second spectral information of each spectral channel and the interval spectral information to obtain the normalized image.

[0160] In the above embodiment of the present application, the device further comprises: an output module configured to output the initial spectral image and the target spectral image; and a receiving module configured to receive feedback information, wherein the feedback information is used to represent whether the spectral noise is superimposed on the initial spectral image; and the identification module is further configured to identify the target spectral image to determine the category of the target object in a case where the feedback information is that the spectral noise is superimposed on the initial spectral image.

[0161] In the above embodiment of the present application, the device further comprises: a determining module configured to determine a target display mode corresponding to the target object based on the category of the target object; and a marking module configured to mark the target object in the target spectral image according to the target display mode.

[0162] It should be noted that the preferred implementation schemes and embodiments involved in the above embodiments of the present application have the same application scenarios and implementation processes as the schemes provided in Embodiment 1, but are not limited to the schemes provided in Embodiment 1.

[0163] Embodiment 5

[0164] According to the embodiments of the present application, an image processing device for implementing the above image processing method is further provided, as shown in Figure 9 The device 900 comprises: an obtaining module 902, a superimposition module 904, and an identification module 906.

[0165] The acquisition module 902 is configured to acquire an initial spectral image, wherein the initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels.

[0166] It should be noted that the acquisition module 902, the superposition module 904, and the identification module 906 correspond to steps S602 to S606 in Embodiment 2, and the three modules have the same instances and application scenarios as the corresponding steps, but are not limited to the content disclosed in Embodiment 2. It should be noted that the modules can run in the computer terminal 10 provided in Embodiment 1 as part of the device.

[0167] It should be noted that the preferred embodiments involved in the above embodiments of the present application have the same scheme, application scenario, and implementation process as Embodiment 1, but are not limited to the scheme provided in Embodiment 1.

[0168] Embodiment 6

[0169] According to the embodiments of the present application, an image processing device for implementing the above image processing method is also provided, as shown in the device 1000 includes a receiving module 1002, a superposition module 1004, and an identification module 1006. Figure 10

[0170] The receiving module 1002 is configured to receive, by a cloud server, an initial spectral image uploaded by a client, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels. The superposition module 1004 is configured to superimpose, by the cloud server, spectral noise on the initial spectral image to obtain a target spectral image. The identification module 1006 is configured to identify, by the cloud server, the target spectral image to determine the category of the target object.

[0171] It should be noted that the receiving module 1002, the superposition module 1004, and the identification module 1006 correspond to steps S702 to S706 in Embodiment 3, and the three modules have the same instances and application scenarios as the corresponding steps, but are not limited to the content disclosed in Embodiment 3. It should be noted that the modules can run in the computer terminal 10 provided in Embodiment 1 as part of the device.

[0172] It should be noted that the preferred embodiments involved in the above embodiments of the present application have the same scheme, application scenario, and implementation process as Embodiment 1, but are not limited to the scheme provided in Embodiment 1. It should be noted that the preferred embodiments involved in the above embodiments of the present application have the same scheme, application scenario, and implementation process as Embodiment 1, but are not limited to the scheme provided in Embodiment 1.

[0173] Embodiment 7

[0174] According to the embodiments of the present application, an image processing method embodiment is further provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from here.

[0175] Figure 11 is a flowchart of an image processing method according to Embodiment 7 of the present application. As shown in the figure, the method can include the following steps: Figure 11

[0176] Step S1102, acquiring an initial spectral image.

[0177] The initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels.

[0178] Step S1104, superimposing the spectral noise on the initial spectral image to obtain a target spectral image.

[0179] Step S1106, identifying the target spectral image to determine the yield of the target crop.

[0180] In the above embodiments of the present application, superimposing the spectral noise on the initial spectral image to obtain a target spectral image includes at least one of the following: superimposing the image information and the spectral noise to obtain the target spectral image; superimposing the first spectral information of the plurality of spectral channels and the spectral noise to obtain the target spectral image.

[0181] In the above embodiments of the present application, superimposing the image information and the spectral noise to obtain the target spectral image includes: obtaining the product of the image information and the noise intensity of the spectral noise to obtain a first product; obtaining the sum of the first product and the spectral noise to obtain the target spectral image.

[0182] In the above embodiments of the present application, superimposing the first spectral information of the plurality of spectral channels and the spectral noise to obtain the target spectral image includes: obtaining the product of the first spectral information of each spectral channel and the noise intensity of the spectral noise to obtain a first product corresponding to each spectral channel; obtaining the sum of the first product corresponding to each spectral channel and the spectral noise to obtain the second spectral information of each spectral channel contained in the target spectral image.

[0183] ​In the above embodiment of the present application, after superimposing the spectral noise on the initial spectral image to obtain the target spectral image, the method further comprises: performing normalization processing on the second spectral information of each spectral channel contained in the target spectral image to obtain a normalized image; and identifying the normalized image to determine the yield of the target crop.

[0184] In the above embodiment of the present application, performing normalization processing on the second spectral information of each spectral channel contained in the target spectral image to obtain a normalized image comprises at least one of: performing normalization processing on the second spectral information of adjacent spectral channels to obtain a normalized image; and performing normalization processing on the second spectral information of spectral channels belonging to the same spectral channel interval to obtain a normalized image.

[0185] In the above embodiment of the present application, performing normalization processing on the second spectral information of adjacent spectral channels to obtain a normalized image comprises: obtaining the difference between the second spectral information of adjacent spectral channels to obtain a spectral difference value; obtaining the sum of the second spectral information of adjacent spectral channels to obtain a spectral sum value; and obtaining the ratio of the spectral difference value and the spectral sum value to obtain a normalized image.

[0186] In the above embodiment of the present application, performing normalization processing on the second spectral information of spectral channels belonging to the same spectral channel interval to obtain a normalized image comprises: determining the target spectral channel interval to which each spectral channel belongs; obtaining the sum of at least one spectral channel corresponding to the target spectral channel interval to obtain interval spectral information; and obtaining the ratio of the second spectral information of each spectral channel and the interval spectral information to obtain a normalized image.

[0187] In the above embodiment of the present application, after superimposing the spectral noise on the initial spectral image to obtain the target spectral image, the method further comprises: outputting the initial spectral image and the target spectral image; receiving feedback information, wherein the feedback information is used to represent whether to superimpose the spectral noise on the initial spectral image; and in the case that the feedback information is to superimpose the spectral noise on the initial spectral image, identifying the target spectral image to determine the yield of the target crop.

[0188] In the above embodiment of the present application, after identifying the target spectral image to determine the yield of the target crop, the method further comprises: determining a target display mode corresponding to the target crop based on the yield of the target crop; and marking the target crop in the target spectral image according to the target display mode.

[0189] It should be noted that the preferred embodiments involved in the above embodiments of the present application have the same application scenarios and implementation processes as the schemes provided in Embodiment 1, but are not limited to the schemes provided in Embodiment 1.

[0190] Embodiment 8

[0191] According to the embodiments of the present application, an image processing device for implementing the image processing method is also provided, as shown in the figure, the device 1200 comprises an acquisition module 1202, a superposition module 1204 and an identification module 1206. Figure 12

[0192] The acquisition module 1202 is configured to acquire an initial spectral image, wherein the initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; the superposition module 1204 is configured to superimpose spectral noise on the initial spectral image to obtain a target spectral image; and the identification module 1206 is configured to identify the target spectral image to determine the yield of the target crop.

[0193] It should be noted that the acquisition module 1202, the superposition module 1204 and the identification module 1206 correspond to steps S1102 to S1106 in Embodiment 7, and the three modules have the same instances and application scenarios as the corresponding steps, but are not limited to the above-mentioned contents disclosed in Embodiment 2. It should be noted that the above-mentioned modules as part of the device can run in the computer terminal 10 provided in Embodiment 1.

[0194] It should be noted that the preferred embodiments involved in the above embodiments of the present application have the same scheme, application scenario and implementation process as the scheme provided in Embodiment 1, but are not limited to the scheme provided in Embodiment 1.

[0195] Embodiment 9

[0196] The embodiments of the present application can provide a computer terminal, which can be any one of the computer terminal devices in the computer terminal group. Alternatively, in the present embodiment, the above-mentioned computer terminal can also be replaced by a terminal device such as a mobile terminal.

[0197] Alternatively, in the present embodiment, the above-mentioned computer terminal can be located in at least one of the network devices in the computer network.

[0198] In the present embodiment, the above-mentioned computer terminal can execute the program code of the following steps in the image processing method: acquiring an initial spectral image, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; and identifying the target spectral image to determine the category of the target object.

[0199] Alternatively, Figure 13 is a structural block diagram of a computer terminal according to the embodiments of the present application. As shown in the figure, Figure 13 ​As shown, the computer terminal 1300 can include one or more (only one shown) processors 1302, memory 1304.

[0200] The memory can be used to store software programs and modules, such as program instructions / modules corresponding to the image processing method and device in the embodiments of the present application. The processor executes various functions and data processing by running the software programs and modules stored in the memory, that is, implements the image processing method described above. The memory can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory can further include a memory remotely arranged with respect to the processor, which can be connected to the terminal 1300 through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0201] The processor can call information and application programs stored in the memory through the transmission device to execute the following steps: obtaining an initial spectral image, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; and identifying the target spectral image to determine the category of the target object.

[0202] Optionally, the processor can further execute program codes of the following steps: superimposing the image information on the spectral noise to obtain the target spectral image; and superimposing the first spectral information of the plurality of spectral channels on the spectral noise to obtain the target spectral image.

[0203] Optionally, the processor can further execute program codes of the following steps: obtaining a product of the image information and the noise intensity of the spectral noise to obtain a first product; and obtaining a sum of the first product and the spectral noise to obtain the target spectral image.

[0204] Optionally, the processor can further execute program codes of the following steps: obtaining a product of the first spectral information of each spectral channel and the noise intensity of the spectral noise to obtain a first product corresponding to each spectral channel; and obtaining a sum of the first product corresponding to each spectral channel and the spectral noise to obtain second spectral information of each spectral channel contained in the target spectral image.

[0205] In the above embodiments of the present application, after superimposing the spectral noise on the initial spectral image to obtain the target spectral image, the method further includes: performing normalization processing on the second spectral information of each spectral channel contained in the target spectral image to obtain a normalized image; and identifying the normalized image to determine the category of the target object.

[0206] Optionally, the processor can further execute program codes of the following steps: normalizing the second spectral information of adjacent spectral channels to obtain a normalized image; normalizing the second spectral information of spectral channels belonging to the same spectral channel interval to obtain a normalized image.

[0207] Optionally, the processor can further execute program codes of the following steps: obtaining the difference between the second spectral information of adjacent spectral channels to obtain a spectral difference value; obtaining the sum of the second spectral information of adjacent spectral channels to obtain a spectral sum value; obtaining the ratio of the spectral difference value and the spectral sum value to obtain a normalized image.

[0208] Optionally, the processor can further execute program codes of the following steps: determining the target spectral channel interval to which each spectral channel belongs; obtaining the sum of at least one spectral channel corresponding to the target spectral channel interval to obtain interval spectral information; obtaining the ratio of the second spectral information of each spectral channel and the interval spectral information to obtain a normalized image.

[0209] Optionally, the processor can further execute program codes of the following steps: after superimposing the spectral noise on the initial spectral image to obtain the target spectral image, outputting the initial spectral image and the target spectral image; receiving feedback information, wherein the feedback information is used to represent whether to superimpose the spectral noise on the initial spectral image; in the case that the feedback information is to superimpose the spectral noise on the initial spectral image, identifying the target spectral image to determine the category of the target object.

[0210] Optionally, the processor can further execute program codes of the following steps: after identifying the target spectral image to determine the category of the target object, determining the target display mode corresponding to the target object based on the category of the target object; and marking the target object in the target spectral image according to the target display mode.

[0211] The processor can call the information and application programs stored in the memory through the transmission device to execute the following steps: obtaining an initial spectral image, wherein the initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; identifying the target spectral image to determine the category of the target crop.

[0212] The processor can call information and application programs stored in the memory through the transmission device to perform the following steps: the cloud server receives an initial spectral image uploaded by a client, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; the cloud server superimposes spectral noise on the initial spectral image to obtain a target spectral image; and the cloud server identifies the target spectral image to determine the category of the target object.

[0213] The processor can call information and application programs stored in the memory through the transmission device to perform the following steps: obtaining an initial spectral image, wherein the initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; and identifying the target spectral image to determine the yield of the target crop.

[0214] Those skilled in the art can understand that, Figure 13 The structure shown is only schematic, and the computer terminal can also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, a Mobile Internet Device (MID), a PAD, or the like. Figure 13 It does not limit the structure of the electronic device. For example, the computer terminal 1100 can further include more or fewer components (such as a network interface, a display device, etc.) than those shown in the figure, or have a different configuration from that shown in the figure. Figure 13 It does not limit the structure of the electronic device. For example, the computer terminal 1100 can further include more or fewer components (such as a network interface, a display device, etc.) than those shown in the figure, or have a different configuration from that shown in the figure. Figure 13 It does not limit the structure of the electronic device. For example, the computer terminal 1100 can further include more or fewer components (such as a network interface, a display device, etc.) than those shown in the figure, or have a different configuration from that shown in the figure.

[0215] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be instructed by a program to instruct the related hardware of the terminal device, and the program can be stored in a computer readable storage medium, which can include a flash disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, etc.

[0216] Embodiment 10

[0217] The embodiments of the present application also provide a storage medium. Optionally, in the embodiment, the above storage medium can be used to save the program code executed by the image processing method provided by the above embodiments.

[0218] Optionally, in the embodiment, the above storage medium can be located in any one of the computer terminals in the computer terminal group in the computer network, or in any one of the mobile terminals in the mobile terminal group.

[0219] Optionally, in the embodiment, the storage medium is configured to store program code for performing the following steps: obtaining an initial spectral image, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; and identifying the target spectral image to determine a category of the target object.

[0220] Optionally, the storage medium is further configured to store program code for performing the following steps: superimposing the image information on the spectral noise to obtain the target spectral image; and superimposing the first spectral information of the plurality of spectral channels on the spectral noise to obtain the target spectral image.

[0221] Optionally, the storage medium is further configured to store program code for performing the following steps: obtaining a product of the image information and a noise intensity of the spectral noise to obtain a first product; and obtaining a sum of the first product and the spectral noise to obtain the target spectral image.

[0222] Optionally, the storage medium is further configured to store program code for performing the following steps: obtaining a product of the first spectral information of each spectral channel and a noise intensity of the spectral noise to obtain a first product corresponding to each spectral channel; and obtaining a sum of the first product corresponding to each spectral channel and the spectral noise to obtain second spectral information of each spectral channel contained in the target spectral image.

[0223] Optionally, the storage medium is further configured to store program code for performing the following steps: performing normalization processing on the second spectral information of each spectral channel contained in the target spectral image to obtain a normalized image; and identifying the normalized image to determine the category of the target object.

[0224] Optionally, the storage medium is further configured to store program code for performing the following steps: performing normalization processing on the second spectral information of adjacent spectral channels to obtain a normalized image; and performing normalization processing on the second spectral information of spectral channels belonging to a same spectral channel interval to obtain a normalized image.

[0225] Optionally, the storage medium is further configured to store program code for performing the following steps: obtaining a difference between the second spectral information of adjacent spectral channels to obtain a spectral difference value; obtaining a sum of the second spectral information of adjacent spectral channels to obtain a spectral sum value; and obtaining a ratio of the spectral difference value and the spectral sum value to obtain a normalized image.

[0226] Optionally, the storage medium is further configured to store program code for performing the following steps: determining a target spectral channel interval to which each spectral channel belongs; obtaining a sum of at least one spectral channel corresponding to the target spectral channel interval to obtain interval spectral information; and obtaining a ratio of second spectral information of each spectral channel to the interval spectral information to obtain a normalized image.

[0227] Optionally, the storage medium is further configured to store program code for performing the following steps: after superimposing the spectral noise on the initial spectral image to obtain the target spectral image, outputting the initial spectral image and the target spectral image; receiving feedback information, wherein the feedback information is used to represent whether to superimpose the spectral noise on the initial spectral image; and in a case where the feedback information is to superimpose the spectral noise on the initial spectral image, identifying the target spectral image to determine a category of the target object.

[0228] Optionally, the storage medium is further configured to store program code for performing the following steps: after identifying the target spectral image to determine the category of the target object, determining a target display mode corresponding to the target object based on the category of the target object; and marking the target object in the target spectral image according to the target display mode.

[0229] Optionally, in the embodiment, the storage medium is configured to store program code for performing the following steps: obtaining an initial spectral image, wherein the initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; and identifying the target spectral image to determine a category of the target crop.

[0230] Optionally, in the embodiment, the storage medium is configured to store program code for performing the following steps: a cloud server receives an initial spectral image uploaded by a client, wherein the initial spectral image contains a target object, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; the cloud server superimposes spectral noise on the initial spectral image to obtain a target spectral image; and the cloud server identifies the target spectral image to determine a category of the target object.

[0231] Optionally, in the embodiment, the storage medium is configured to store program code for performing the following steps: obtaining an initial spectral image, wherein the initial spectral image contains a target crop, and the initial spectral image is composed of image information and first spectral information of a plurality of spectral channels; superimposing spectral noise on the initial spectral image to obtain a target spectral image; and identifying the target spectral image to determine a yield of the target crop.

[0232] The serial numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0233] In the above-described embodiments of the present application, the description of each embodiment focuses on different aspects, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0234] In several embodiments provided in the present application, it should be understood that the disclosed technical contents can be implemented by other ways. Among them, the above-described device embodiments are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division way, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, which can be electrical or other forms.

[0235] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0236] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The above integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0237] The integrated unit, if realized in the form of software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part of the prior art or the whole or part of the technical solutions can be embodied in the form of software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and various program codes that can be stored in the medium.

[0238] The above merely describes the preferred embodiments of the present application, and it should be pointed out that, for those skilled in the art, some improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. An image processing method, characterized in that, include: Acquire an initial spectral image, wherein the initial spectral image contains the target object, and the initial spectral image is composed of image information and first spectral information of multiple spectral channels; The spectral noise is superimposed on the initial spectral image to obtain the target spectral image; The target spectral image is identified to determine the category of the target object; The method of superimposing the spectral noise with the initial spectral image to obtain the target spectral image includes at least one of the following: superimposing the image information with the spectral noise to obtain the target spectral image; or superimposing the first spectral information of the plurality of spectral channels with the spectral noise to obtain the target spectral image.

2. The method according to claim 1, characterized in that, The image information is superimposed with the spectral noise to obtain the target spectral image, including: The first product is obtained by multiplying the image information by the noise intensity of the spectral noise. The target spectral image is obtained by summing the first product with the spectral noise.

3. The method according to claim 1, characterized in that, The target spectral image is obtained by superimposing the first spectral information of the multiple spectral channels with the spectral noise, including: The product of the first spectral information of each spectral channel and the noise intensity of the spectral noise is obtained to obtain the first product corresponding to each spectral channel; The sum of the first product corresponding to each spectral channel and the spectral noise is obtained to obtain the second spectral information of each spectral channel contained in the target spectral image.

4. The method according to any one of claims 1 to 3, characterized in that, After superimposing the spectral noise onto the initial spectral image to obtain the target spectral image, the method further includes: The second spectral information of each spectral channel contained in the target spectral image is normalized to obtain a normalized image; The normalized image is identified to determine the category of the target object.

5. The method according to claim 4, characterized in that, The second spectral information of each spectral channel contained in the target spectral image is normalized to obtain the normalized image, which includes at least one of the following: The second spectral information of adjacent spectral channels is normalized to obtain the normalized image; The second spectral information of spectral channels belonging to the same spectral channel range is normalized to obtain the normalized image.

6. The method according to claim 1, characterized in that, After superimposing the spectral noise onto the initial spectral image to obtain the target spectral image, the method further includes: Output the initial spectral image and the target spectral image; Receive feedback information, wherein the feedback information is used to characterize whether to superimpose the spectral noise onto the initial spectral image; When the feedback information is to superimpose the spectral noise onto the initial spectral image, the target spectral image is identified to determine the category of the target object.

7. The method according to claim 1, characterized in that, After identifying the target spectral image and determining the category of the target object, the method further includes: Based on the category of the target object, determine the target display method corresponding to the target object; According to the target display method, the target object is marked in the target spectral image.

8. An image processing method, characterized in that, include: Acquire an initial spectral image, wherein the initial spectral image contains the target crop, and the initial spectral image is composed of image information and first spectral information of multiple spectral channels; The spectral noise is superimposed on the initial spectral image to obtain the target spectral image; The target spectral image is identified to determine the category of the target crop; The method of superimposing the spectral noise with the initial spectral image to obtain the target spectral image includes at least one of the following: superimposing the image information with the spectral noise to obtain the target spectral image; or superimposing the first spectral information of the plurality of spectral channels with the spectral noise to obtain the target spectral image.

9. An image processing method, characterized in that, include: Acquire an initial spectral image, wherein the initial spectral image contains the target crop, and the initial spectral image is composed of image information and first spectral information of multiple spectral channels; The spectral noise is superimposed on the initial spectral image to obtain the target spectral image; The target spectral image is identified to determine the yield of the target crop; The method of superimposing the spectral noise with the initial spectral image to obtain the target spectral image includes at least one of the following: superimposing the image information with the spectral noise to obtain the target spectral image; or superimposing the first spectral information of the plurality of spectral channels with the spectral noise to obtain the target spectral image.

10. An image processing method, characterized in that, include: The cloud server receives an initial spectral image uploaded by the client, wherein the initial spectral image contains the target object and is composed of image information and first spectral information of multiple spectral channels; The cloud server superimposes the spectral noise onto the initial spectral image to obtain the target spectral image; The cloud server identifies the target spectral image to determine the category of the target object; The method of superimposing the spectral noise with the initial spectral image to obtain the target spectral image includes at least one of the following: superimposing the image information with the spectral noise to obtain the target spectral image; or superimposing the first spectral information of the plurality of spectral channels with the spectral noise to obtain the target spectral image.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the image processing method according to any one of claims 1 to 9.

12. A computer terminal, characterized in that, include: A memory and a processor, the processor being configured to run a program stored in the memory, wherein the program, when executed, performs the image processing method according to any one of claims 1 to 9.

Images