An image segmentation method, device, electronic equipment and storage medium

Abstract

The application discloses an image segmentation method and device, electronic equipment and storage medium. Including: obtaining image data to be segmented, segmenting the image data to be segmented based on preset channel threshold data to obtain an initial segmentation result including at least one initial segmentation region; determining the prompt point information corresponding to each initial segmentation region, inputting the prompt point information and the image data to be segmented into a preset image segmentation model for processing to obtain a first image segmentation result of the image data to be segmented; obtaining preset text prompt information, inputting the preset text prompt information and the image data to be segmented into the preset image segmentation model for processing to obtain a second image segmentation result of the image data to be segmented; determining a target image segmentation result of the image data to be segmented based on the similarity data of the first image segmentation result and the second image segmentation result. The segmentation result can be determined without additional annotation data, improving the high precision and adaptability of image segmentation.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to an image segmentation method, apparatus, electronic device, and storage medium. Background Technology

[0002] As a critical component of the power system, the safe and stable operation of substations is essential for ensuring the overall efficiency of the power grid. After prolonged use, equipment in substations may develop problems such as rust and cracks. Rust significantly reduces the insulation performance of the equipment, posing a considerable risk of malfunction and disruption to normal operation. Cracks also pose safety hazards.

[0003] Traditional methods for inspecting substation equipment rely on manual inspection, which becomes increasingly inefficient as the number of substations grows. Image processing methods can improve efficiency by segmenting problematic areas, but these methods suffer from inaccurate segmentation results. While neural network models like U-Net, SegNet, and DeepLab can be used for image segmentation, they require extensive labeled data for training. Furthermore, if the segmented object changes, the model needs to be retrained, resulting in high computational resource consumption and poor adaptability. Summary of the Invention

[0004] This invention provides an image segmentation method, apparatus, electronic device, and storage medium to solve the problems of existing image segmentation technologies, such as the need for large amounts of standard data, high computational resource consumption, and poor adaptability.

[0005] According to one aspect of the present invention, an image segmentation method is provided, comprising:

[0006] Acquire image data to be segmented, and perform segmentation processing on the image data to be segmented based on preset channel threshold data to obtain an initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region;

[0007] Determine the cue point information corresponding to each initial segmentation region, input the cue point information and the image data to be segmented into the preset image segmentation model for processing, and obtain the first image segmentation result of the image data to be segmented. The first image segmentation result is used to characterize the target segmentation region in the image data to be segmented.

[0008] Obtain preset text prompt information, input the preset text prompt information and the image data to be segmented into the preset image segmentation model for processing, and obtain the second image segmentation result of the image data to be segmented. The preset text prompt information is the prompt information of the object to which the target segmentation region belongs, and the second image segmentation result is used to characterize the segmentation region where the object to which the target segmentation region belongs is located in the image data to be segmented.

[0009] The similarity data between the first image segmentation result and the second image segmentation result is determined, and the target image segmentation result of the image data to be segmented is determined based on the similarity data.

[0010] Optionally, the preset channel threshold data is preset tone threshold data; the image data to be segmented is segmented based on the preset channel threshold data to obtain an initial segmentation result, including: performing color space conversion on the image data to be segmented to obtain image data in a preset color space, the image data in the preset color space including tone channel data; comparing the tone channel data of the image data to be segmented with the preset tone threshold data, segmenting out the regions that meet the preset tone threshold data, and obtaining the initial segmentation result.

[0011] Optionally, before determining the cue point information corresponding to each initial segmentation region, the method includes: performing image preprocessing on the initial segmentation result to obtain a processed initial segmentation result, wherein the image preprocessing includes morphological processing and filtering processing, and the initial segmentation result includes at least one connected component.

[0012] Optionally, the prompt point information and the image data to be segmented are input into a preset image segmentation model for processing to obtain the first image segmentation result of the image data to be segmented, including: traversing each initial segmentation region, inputting the prompt point information and the image data to be segmented of each initial segmentation region into the preset image segmentation model for processing to obtain the first sub-segmentation result corresponding to each initial segmentation region, and the first sub-segmentation results forming the first image segmentation result.

[0013] Optionally, determining the similarity data between the first image segmentation result and the second image segmentation result, and determining the target image segmentation result of the image data to be segmented based on the similarity data, includes: determining the similarity data between the first image segmentation result and the second image segmentation result; comparing the similarity data with a preset similarity threshold, and if the similarity data is less than the preset similarity threshold, then taking the first image segmentation result as the target image segmentation result.

[0014] Optionally, the first image segmentation result includes at least one first sub-segmentation result corresponding to at least one target segmentation region, and the second image segmentation result includes at least one second sub-segmentation result corresponding to at least one object segmentation region; determining the similarity data between the first image segmentation result and the second image segmentation result includes: traversing the first image segmentation result and the second image segmentation result, and determining the similarity data between any first sub-segmentation result and any second sub-segmentation result.

[0015] Optionally, determining the similarity data between the first image segmentation result and the second image segmentation result, and determining the target image segmentation result of the image data to be segmented based on the similarity data, further includes: if any similarity data is greater than or equal to a preset similarity threshold, then determining the target segmentation region corresponding to the similarity data greater than or equal to the preset similarity threshold, adjusting the prompt point information of the initial segmentation result corresponding to the target segmentation region, and returning to execute the step of inputting the prompt point information and the image data to be segmented into a preset image segmentation model for processing based on the adjusted prompt point information, so as to obtain the first image segmentation result of the image data to be segmented, thereby obtaining the target image segmentation result.

[0016] According to another aspect of the present invention, an image segmentation apparatus is provided, comprising:

[0017] The initial segmentation result determination module is used to acquire the image data to be segmented, perform segmentation processing on the image data to be segmented based on preset channel threshold data, and obtain an initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region.

[0018] The first image segmentation result determination module is used to determine the prompt point information corresponding to each initial segmentation region, input the prompt point information and the image data to be segmented into the preset image segmentation model for processing, and obtain the first image segmentation result of the image data to be segmented. The first image segmentation result is used to characterize the target segmentation region in the image data to be segmented.

[0019] The second image segmentation result determination module is used to obtain preset text prompt information, input the preset text prompt information and the image data to be segmented into the preset image segmentation model for processing, and obtain the second image segmentation result of the image data to be segmented. The preset text prompt information is the prompt information of the object to which the target segmentation region belongs, and the second image segmentation result is used to characterize the segmentation region where the object to which the target segmentation region belongs is located in the image data to be segmented.

[0020] The target image segmentation result determination module is used to determine the similarity data between the first image segmentation result and the second image segmentation result, and to determine the target image segmentation result of the image data to be segmented based on the similarity data.

[0021] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0022] At least one processor; and

[0023] A memory that is communicatively connected to at least one processor; wherein,

[0024] The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the image segmentation method of any embodiment of the present invention.

[0025] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement the image segmentation method of any embodiment of the present invention.

[0026] The technical solution of this invention involves acquiring image data to be segmented, segmenting the image data based on preset channel threshold data to obtain an initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region; determining prompt point information corresponding to each initial segmentation region, and inputting the prompt point information and the image data to be segmented into a preset image segmentation model for processing to obtain a first image segmentation result of the image data to be segmented, wherein the first image segmentation result is used to characterize the target segmentation region in the image data to be segmented; acquiring preset text prompt information, and inputting the preset text prompt information and the image data to be segmented into the preset image segmentation model for processing to obtain a second image segmentation result of the image data to be segmented. The system employs a pre-defined text prompt message, which serves as a prompt message for the object to which the target segmentation region belongs. The second image segmentation result is used to characterize the segmentation region where the object to which the target segmentation region belongs is located in the image data to be segmented. The similarity data between the first and second image segmentation results is determined, and the target image segmentation result of the image data to be segmented is determined based on the similarity data. This solves the problem of needing a large amount of labeled data to determine the segmentation of image data, reduces the dependence on a large amount of labeled data, eliminates the need for additional model training, and enhances the image segmentation model's ability to recognize images in complex environments by adopting a prompt strategy. This improves the speed, accuracy, and applicability of segmenting the target segmentation region in the image data to be segmented.

[0027] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart of an image segmentation method provided in Embodiment 1 of the present invention;

[0030] Figure 2 This is a flowchart of an image segmentation method provided in Embodiment 2 of the present invention;

[0031] Figure 3 This is a schematic diagram of the structure of an image segmentation device provided in Embodiment 3 of the present invention;

[0032] Figure 4 This is a schematic diagram of the structure of an electronic device that implements the image segmentation method of this invention. Detailed Implementation

[0033] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0034] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0035] Example 1

[0036] Figure 1 This is a flowchart of an image segmentation method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations involving image segmentation. The method can be executed by an image segmentation device, which can be implemented in hardware and / or software. This image segmentation device can be configured in electronic devices such as servers and computers. Figure 1 As shown, the method includes:

[0037] S110. Obtain the image data to be segmented, and perform segmentation processing on the image data to be segmented based on the preset channel threshold data to obtain the initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region.

[0038] Specifically, the image data to be segmented refers to the image data that requires image segmentation processing. This data can be acquired periodically or in real-time using an image acquisition device. The acquisition method can be set according to the actual needs of image segmentation. It should be noted that image acquisition devices can be installed in any scenario requiring inspection, and the acquired image data can be used as the image data to be segmented. For example, in a substation scenario, an image acquisition device can be installed in the environment requiring inspection to acquire image data. This image data includes at least one object to be inspected, which can be an instrument or piece of equipment. For example, it can be used to check the integrity of the instrument or equipment, or to check for rust. The preset channel threshold data can be understood as data set for coarse segmentation of the image data to be segmented, initially segmenting the region of interest in the image data. Specifically, this refers to the color channel data in the color space. For example, in the HSV color space, the preset channel threshold data can be the threshold data corresponding to one of the hue channel, saturation channel, and luminance channel. The preset channel threshold data is selected and set according to actual needs. It should be noted that the preset channel threshold data are different in different color spaces. For determining a reasonable color space for the image data to be segmented, it can be determined based on the characteristics of the target segmentation region in the image data. Features can include color features, material properties, and shape features, etc. For example, taking the rusted area on the device as the target segmentation region, the corresponding channel threshold range can be determined based on the material properties of the device and the color characteristics of the rust on the corresponding material, and the corresponding preset channel threshold data can be obtained. This data is used to coarsely segment the rusted area in the image, that is, to coarsely segment the rusted area that meets the preset channel threshold data. The preset channel threshold data for different target segmentation regions can be set and stored in advance. When needed, the corresponding preset channel threshold data can be matched according to the target segmentation task. The target segmentation task can include the feature data corresponding to the target segmentation region.

[0039] Specifically, image data is acquired in real time by an image acquisition device in the scene to be detected. This acquired image data is used as the image data to be segmented. Based on the image segmentation requirements, preset channel threshold data corresponding to the image data to be segmented is matched from a preset storage space. The image segmentation requirements include feature data of the target segmentation region in the image data to be segmented. Channel data that matches the preset channel threshold data is identified. The channel data of the image data to be segmented is then compared with the preset channel threshold data, and the regions corresponding to the number of channels that meet the preset channel threshold data are segmented, yielding the initial segmentation result. It should be noted that if the color space of the image data to be segmented is different from the color space of the preset channel threshold data, the color space of the image data to be segmented needs to be converted to the color space of the preset channel threshold data to ensure the accuracy of the comparison result between the channel data of the image data to be segmented and the preset channel threshold data.

[0040] In this embodiment, the image data to be segmented is processed by using preset channel threshold data to obtain an initial segmentation result. This achieves pre-segmentation of the image data to be segmented, determines the location of the region of interest in the image data to be segmented, facilitates the rapid location of the target segmentation region for accurate image segmentation, and improves the speed of image segmentation.

[0041] Optionally, the preset channel threshold data is preset tone threshold data; the image data to be segmented is segmented based on the preset channel threshold data to obtain an initial segmentation result, including: performing color space conversion on the image data to be segmented to obtain image data in a preset color space, the image data in the preset color space including tone channel data; comparing the tone channel data of the image data to be segmented with the preset tone threshold data, segmenting out the regions that meet the preset tone threshold data, and obtaining the initial segmentation result.

[0042] In this context, the color channel data corresponding to the image data to be segmented differs in different color spaces. Typically, the color space corresponding to the image data to be segmented is the RGB color space, which obtains various colors by varying the red (R), green (G), and blue (B) color channels and superimposing them. RGB represents the colors of the red, green, and blue channels, but it does not clearly display hue characteristics. The HSV color space can intuitively express the brightness, hue, and vividness of colors. H represents hue, indicating the color category, with a value range of 0° to 360°, used to distinguish different colors. Red corresponds to 0°, green to 120°, and blue to 240°. S represents saturation, indicating the vividness of the color, also known as purity, with a value range of 0% to 100%. Pure colors have a saturation of 100%, while gray has a saturation of 0%. V represents value, indicating the brightness of the color. Black is represented by a brightness of 0%, and white is represented by a brightness of 100%. Values ​​between 0% and 100% represent the lightness or darkness of different colors. The HSV color space facilitates the comparison between colors, especially since the tone channels have significant characteristics. It can be used to quickly determine the region of interest in the image data to be segmented. The data range corresponding to the tone channel data of the region of interest can be set as the preset channel threshold data.

[0043] Specifically, when the preset channel threshold data is the preset hue threshold data, the image data to be segmented is converted to color space, converting the RGB color space of the image data to be segmented to the HSV color space to obtain the hue channel data corresponding to the image data to be segmented. The hue channel data corresponding to the image data to be segmented is compared with the preset channel threshold data, and the image region corresponding to the hue channel data that meets the preset channel threshold data is segmented to obtain the initial segmentation result.

[0044] In this embodiment, leveraging the significant characteristics of the hue channels in the HSV color space, coarse segmentation is performed on the image data to be segmented based on the hue channel data to obtain an initial image segmentation result. This allows for rapid further precise segmentation based on the coarse segmentation result, improving the segmentation speed. It should be noted that the preset channel threshold data can be set according to actual segmentation needs, improving the adaptability of image segmentation.

[0045] S120. Determine the prompting point information corresponding to each initial segmentation region, input the prompting point information and the image data to be segmented into the preset image segmentation model for processing, and obtain the first image segmentation result of the image data to be segmented. The first image segmentation result is used to characterize the target segmentation region in the image data to be segmented.

[0046] Specifically, the cue point information can be understood as multiple pixel points selected from the initial segmentation region. This information includes the number of cue points and their location data. The number of cue points can be set according to actual needs; preferably, it is set to 5. The cue point location data specifically refers to pixel-level location data. The preset image segmentation model can be understood as a neural network model; preferably, it adopts the SAM model. The SAM model can be guided by the cue point information to perform image processing and obtain the image segmentation result, which helps improve the accuracy and efficiency of image segmentation. The target segmentation region can be understood as the region of interest in the image data to be segmented.

[0047] Specifically, a preset number of cue points are selected in each initial segmentation region by random selection, and the position data corresponding to each selected cue point is obtained to obtain the cue point information corresponding to each initial segmentation region. The cue point information corresponding to each initial segmentation region and the image data to be segmented are input as a set of input parameters to the preset image segmentation model. The preset image segmentation model processes each set of input parameters and segments the image data to be segmented according to the cue point information to obtain the first image segmentation result of the image data to be segmented.

[0048] In this embodiment, the prompt information of the coarse segmentation result obtained by the image thresholding segmentation method is input together with the image data to be segmented into the preset image segmentation model. The prompt information guides the model to process the image data to be segmented, so that the segmentation result matching the prompt information can be obtained quickly, thereby improving the speed and accuracy of image segmentation.

[0049] Optionally, the prompt point information and the image data to be segmented are input into a preset image segmentation model for processing to obtain the first image segmentation result of the image data to be segmented, including: traversing each initial segmentation region, inputting the prompt point information and the image data to be segmented of each initial segmentation region into the preset image segmentation model for processing to obtain the first sub-segmentation result corresponding to each initial segmentation region, and the first sub-segmentation results forming the first image segmentation result.

[0050] Specifically, each initial segmentation region is traversed, and each initial segmentation region and the image data to be segmented are used as a set of input parameters. These are then input into the preset image segmentation model one by one. The preset image segmentation model processes each set of input parameters to obtain the segmentation result corresponding to each set of input parameters, which is the first sub-segmentation result corresponding to each initial segmentation region. The set of all the first sub-segmentation results corresponding to the initial segmentation regions is used as the first image segmentation result.

[0051] Optionally, before determining the cue point information corresponding to each initial segmentation region, the method includes: performing image preprocessing on the initial segmentation result to obtain a processed initial segmentation result, wherein the image preprocessing includes morphological processing and filtering processing, and the initial segmentation result includes at least one connected component.

[0052] Image preprocessing specifically refers to preprocessing the image according to a preset image preprocessing method to remove false detection regions in the image, thereby improving the accuracy of subsequent image segmentation results. Image preprocessing includes morphological processing and filtering. Morphological processing refers to performing erosion-dilation processing on the image; filtering refers to filtering out regions of interest (ROIs) from the image. ROIs can be identified and labeled using image annotation methods to obtain their size and aspect ratio characteristics, and then filtering is performed based on these labeled size and aspect ratio characteristics.

[0053] Specifically, each initially segmented region in the initial segmentation result undergoes an erosion operation to remove small noise points. Then, a dilation process connects adjacent regions with a distance less than a preset distance threshold, forming continuous regions, i.e., the corresponding connected components. Through erosion and dilation, each connected component in the initial segmentation result is obtained. The initial segmentation result includes at least one connected component. It should be noted that if the initial segmentation result has no connected components, no further processing is needed, and the segmentation result can be directly output. After obtaining each connected component in the initial segmentation result, an image annotation algorithm is called to identify and annotate each connected component, obtaining the size and aspect ratio features of each connected component. It is then determined whether the size and aspect ratio features of each connected component meet the corresponding preset size threshold and aspect ratio threshold. If the size or aspect ratio feature of any connected component does not meet the corresponding preset size threshold, the connected component is deleted.

[0054] In this embodiment, the initial segmentation results are processed through morphological processing and filtering to remove false detection areas, small shadow areas and highlight points, as well as areas that do not require attention, thereby reducing the amount of data processing and improving the speed and accuracy of image segmentation.

[0055] S130. Obtain preset text prompt information, input the preset text prompt information and the image data to be segmented into a preset image segmentation model for processing, and obtain the second image segmentation result of the image data to be segmented. The preset text prompt information is the prompt information of the object to which the target segmentation region belongs, and the second image segmentation result is used to characterize the segmentation region where the object to which the target segmentation region belongs is located in the image data to be segmented.

[0056] The preset text prompt information can be understood as information representing the object to be segmented in the image data. This preset text prompt information can be pre-configured in the configuration information. When it is needed, the preset text prompt information matching the segmented object can be read from the configuration information. In this embodiment, the preset text prompt information is the prompt information for the object to which the target segmentation region belongs. For example, in a substation, to specify segmenting instruments and equipment in the substation image data, the preset text prompt information can be set to "Substation Related Equipment," or to specify segmenting only transformers in the substation image data, the preset text prompt information can be set to "Transformer of the Substation." The preset text prompt information can be pre-set according to actual needs. For example, if the target segmentation requirement is to segment the rusted area on the equipment in the substation, the rusted area is taken as the target segmentation region, and the equipment corresponding to the segmented target segmentation region is the segmentation region containing the object to which the target segmentation region belongs.

[0057] Specifically, preset text prompts can be obtained from the configuration information, a preset image segmentation model can be called, the preset text prompts and the image data to be segmented can be input into the preset image segmentation model, and the preset image segmentation model can perform segmentation processing on the image data to be segmented according to the preset text prompts, segmenting the device region in the image data to be segmented, and obtaining the second image segmentation result of the image data to be segmented.

[0058] Based on the above embodiments, a preset image segmentation model is used to segment the image data to be segmented according to preset text prompts. The second image segmentation result of the image data to be segmented is used to characterize the segmentation region of the target segmentation region in the image data to be segmented. The preset image segmentation model can not only separate the segmentation region of the object and generate the corresponding mask image, for example, if the object is an instrument or equipment, then the generated mask image of the equipment region, i.e., the equipment segmentation map, but also, once the segmentation result is determined, a corresponding number is set for each segmented equipment region. The corresponding number set for each equipment region in the same image to be segmented is unique and can be used to determine the equipment number to which each segmentation region in the first image segmentation result belongs. That is, it can not only know the target segmentation region segmented in the image data to be segmented, but also know the object number to which the target segmentation region belongs. For example, if it is to detect whether there is a rusted area in the equipment, then the equipment number corresponding to the rusted area can be determined according to the image segmentation result. This has high reference value for quickly locating equipment with fault risk.

[0059] In this embodiment, a preset image segmentation model is guided to perform image segmentation based on preset text prompts to obtain a second image segmentation result. The object to which the target segmentation region belongs is segmented out for subsequent verification of the accuracy of the target segmentation region, so as to ensure the accuracy of the target segmentation region.

[0060] S140. Determine the similarity data between the first image segmentation result and the second image segmentation result, and determine the target image segmentation result of the image data to be segmented based on the similarity data.

[0061] The similarity data represents the degree of similarity between the first image segmentation result and the second image segmentation result, and is used to verify the first image segmentation result. The similarity data can be calculated using image similarity algorithms, including but not limited to Euclidean distance, cosine similarity, and Hausdorff distance algorithms.

[0062] Specifically, by calculating the similarity data between the first image segmentation result and the second image segmentation result using a preset similarity algorithm, the segmented regions in the first image segmentation result can be merged into a group of image data. The similarity data between the merged image data and the second image segmentation result can then be calculated. Alternatively, the similarity data between the segmented regions in the first image segmentation result and the second image segmentation result can be calculated separately, depending on the actual situation. After determining the similarity data, it can be compared with a preset similarity threshold. Based on the comparison result, the target image segmentation result for the image data to be segmented can be determined. For example, if all similarity data meet the preset similarity threshold, the first image segmentation result can be used as the target image segmentation result.

[0063] For example, the segmented regions in the second image segmentation result are extracted, and the similarity data between the segmented regions in the second image segmentation result and each initial segmented region in the first image segmentation result is calculated using the Hausdorff distance algorithm. It is then determined whether the similarity data meets the preset distance threshold data. If none of them meet the threshold, it means that the similarity between the first image segmentation result and the second image segmentation result is low. This indicates that the target segmentation region was not mistakenly segmented into the object region to which the target segmentation region belongs when performing image segmentation using the prompt point information. In this case, the first image segmentation result can be used as the target segmentation result.

[0064] In this embodiment, the accuracy of the image segmentation result is verified by the similarity data of the first image segmentation result and the second image segmentation result. This avoids missegmenting the target segmentation region into the object region to which the target segmentation region belongs when performing image segmentation based on the prompt point information, thereby improving the accuracy of the image segmentation result.

[0065] The technical solution of this embodiment involves acquiring image data to be segmented, segmenting the image data based on preset channel threshold data to obtain an initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region; determining the prompt point information corresponding to each initial segmentation region, and inputting the prompt point information and the image data to be segmented into a preset image segmentation model for processing to obtain a first image segmentation result of the image data to be segmented, which is used to characterize the target segmentation region in the image data to be segmented; acquiring preset text prompt information, and inputting the preset text prompt information and the image data to be segmented into the preset image segmentation model for processing to obtain a second image segmentation result of the image data to be segmented. The system employs a pre-defined text prompt message, which serves as a prompt message for the object to which the target segmentation region belongs. The second image segmentation result is used to characterize the segmentation region in the image data to be segmented, where the object to which the target segmentation region belongs is located. The similarity data between the first and second image segmentation results is determined, and the target image segmentation result of the image data to be segmented is determined based on the similarity data. This solves the problem of requiring a large amount of labeled data to determine the segmentation of image data, reduces the dependence on a large amount of labeled data, eliminates the need for additional model training, and enhances the image segmentation model's ability to recognize images in complex environments by adopting a prompt strategy. This improves the speed, accuracy, and applicability of segmenting the target segmentation region in the image data to be segmented.

[0066] Example 2

[0067] Figure 2 This is a flowchart of an image segmentation method provided in Embodiment 2 of the present invention. This embodiment is a further optimization of the method in the above embodiment. Optionally, similarity data is compared with a preset similarity threshold. If the similarity data is less than the preset similarity threshold, the first image segmentation result is used as the target image segmentation result. If any similarity data is greater than or equal to the preset similarity threshold, the target segmentation region corresponding to the similarity data greater than or equal to the preset similarity threshold is determined. The prompt point information of the initial segmentation result corresponding to the target segmentation region is adjusted. Based on the adjusted prompt point information, the step of inputting the prompt point information and the image data to be segmented into a preset image segmentation model for processing is returned to obtain the first image segmentation result of the image data to be segmented, so as to obtain the target image segmentation result. Figure 2 As shown, the method includes:

[0068] S210. Obtain the image data to be segmented, and perform segmentation processing on the image data to be segmented based on the preset channel threshold data to obtain the initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region.

[0069] S220. Determine the prompting point information corresponding to each initial segmentation region, input the prompting point information and the image data to be segmented into the preset image segmentation model for processing, and obtain the first image segmentation result of the image data to be segmented. The first image segmentation result is used to characterize the target segmentation region in the image data to be segmented.

[0070] S230. Obtain preset text prompt information, input the preset text prompt information and the image data to be segmented into a preset image segmentation model for processing, and obtain the second image segmentation result of the image data to be segmented. The preset text prompt information is the prompt information of the object to which the target segmentation region belongs, and the second image segmentation result is used to characterize the segmentation region where the object to which the target segmentation region belongs is located in the image data to be segmented.

[0071] S240. Determine the similarity data between the first image segmentation result and the second image segmentation result.

[0072] S250. Compare the similarity data with the preset similarity threshold. If the similarity data is less than the preset similarity threshold, then use the first image segmentation result as the target image segmentation result.

[0073] The preset similarity threshold is set to determine the target image segmentation result. Specifically, it determines the target image segmentation result by judging whether the segmentation result of the first image meets the preset similarity threshold.

[0074] Specifically, the similarity data is compared with a preset similarity threshold to determine whether all similarity data are less than the preset similarity threshold. If all similarity data are less than the preset similarity threshold, it indicates that the target segmentation region has not been segmented into the object region corresponding to the target segmentation region. That is, the first image segmentation result meets the image segmentation requirements, and the first image segmentation result is used as the target image segmentation result. Otherwise, the first image segmentation result needs to be adjusted.

[0075] Based on the above embodiments, the first image segmentation result includes at least one first sub-segmentation result corresponding to at least one target segmentation region, and the second image segmentation result includes at least one second sub-segmentation result corresponding to at least one object segmentation region; determining the similarity data between the first image segmentation result and the second image segmentation result includes: traversing the first image segmentation result and the second image segmentation result to determine the similarity data between any first sub-segmentation result and any second sub-segmentation result.

[0076] Understandably, when the first image segmentation result includes at least one first sub-segmentation result corresponding to at least one target segmentation region, and the second image segmentation result includes at least one second sub-segmentation result corresponding to at least one object segmentation region, it will be necessary to calculate the similarity data between multiple first sub-segmentation results in the first image segmentation result and multiple second sub-segmentation results in the second image segmentation result. In this case, it is necessary to traverse the first and second image segmentation results, extract the multiple second sub-segmentation results in the second image segmentation result, and compare their similarity with the multiple first sub-segmentation results in the first image segmentation result. By using a preset similarity algorithm, pairwise similarity calculations are performed on the multiple first sub-segmentation results in the first image segmentation result and the multiple second sub-segmentation results in the second image segmentation result to determine the similarity data between any first sub-segmentation result and any second sub-segmentation result, thus obtaining multiple similarity data.

[0077] In this embodiment, by determining the similarity data between any first sub-segmentation result and any second sub-segmentation result, the first sub-segmentation result corresponding to similarity data not less than a preset similarity threshold can be determined based on the similarity data determination result. This helps to determine the initial segmentation region to be adjusted. Only the initial segmentation region to be adjusted is re-segmented, eliminating the need to adjust all the initial segmentation regions in the initial segmentation results of the image data to be segmented. This can reduce the workload to a certain extent and help improve the efficiency of image segmentation.

[0078] S260. If any similarity data is greater than or equal to a preset similarity threshold, then determine the target segmentation region corresponding to the similarity data that is greater than or equal to the preset similarity threshold, adjust the prompt point information of the initial segmentation result corresponding to the target segmentation region, and return to execute the step of inputting the prompt point information and the image data to be segmented into the preset image segmentation model for processing to obtain the first image segmentation result of the image data to be segmented, so as to obtain the target image segmentation result.

[0079] Specifically, if any similarity data is greater than or equal to a preset similarity threshold, it is determined that the target segmentation region in the first image segmentation result contains regions that have been missegmented into objects belonging to the target segmentation region. Therefore, the corresponding initial segmentation result needs to be determined based on the target segmentation region corresponding to the similarity data greater than or equal to the preset similarity threshold. Then, the cue point information of this initial segmentation result is adjusted. For example, the number and position of cue points are adjusted. This can be done by randomly selecting cue points of a preset threshold value (or selecting a number greater than the preset threshold value), reading the position data of the corresponding cue points, obtaining the adjusted cue point information, and then re-calling the preset image segmentation model. The adjusted cue point information and the image data to be segmented are used as input data to the preset image segmentation model. The adjusted cue point information guides the preset image segmentation model to perform image segmentation processing on the image data to be segmented, resulting in a new first image segmentation result, which is used to redetermine the target image segmentation result.

[0080] The technical solution of this embodiment involves acquiring image data to be segmented, segmenting the image data based on preset channel threshold data to obtain an initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region; determining the prompt point information corresponding to each initial segmentation region, inputting the prompt point information and the image data to be segmented into a preset image segmentation model for processing to obtain a first image segmentation result of the image data to be segmented, wherein the first image segmentation result is used to characterize the target segmentation region in the image data to be segmented; acquiring preset text prompt information, inputting the preset text prompt information and the image data to be segmented into the preset image segmentation model for processing to obtain a second image segmentation result of the image data to be segmented, wherein the preset text prompt information is the prompt information of the object to which the target segmentation region belongs, and the second image segmentation result is used to characterize the segmentation region in which the object to which the target segmentation region belongs is located in the image data to be segmented; determining the similarity data between the first image segmentation result and the second image segmentation result; comparing the similarity data with a preset similarity threshold, and if the similarity data is less than the preset similarity threshold, then the first image segmentation result is taken as the target image segmentation result. This approach solves the problem of requiring a large amount of labeled data to determine image segmentation, reducing reliance on such data and eliminating the need for additional model training. It employs a cueing strategy to enhance the image segmentation model's ability to recognize images in complex environments, improving the speed and accuracy of target segmentation regions in the image data to be segmented. Furthermore, by adjusting the cue point information corresponding to the initial segmentation regions in the first image segmentation result that does not meet segmentation requirements, the target image segmentation result is redefined. This avoids missegmenting the target segmentation region into its corresponding object region when segmenting the image using cue point information, thus improving the accuracy and applicability of the image segmentation results.

[0081] Example 3

[0082] Figure 3 This is a schematic diagram of the structure of an image segmentation device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes:

[0083] The initial segmentation result determination module 310 is used to acquire the image data to be segmented, perform segmentation processing on the image data to be segmented based on preset channel threshold data, and obtain an initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region.

[0084] The first image segmentation result determination module 320 is used to determine the prompt point information corresponding to each initial segmentation region, input the prompt point information and the image data to be segmented into the preset image segmentation model for processing, and obtain the first image segmentation result of the image data to be segmented. The first image segmentation result is used to characterize the target segmentation region in the image data to be segmented.

[0085] The second image segmentation result determination module 330 is used to obtain preset text prompt information, input the preset text prompt information and the image data to be segmented into a preset image segmentation model for processing, and obtain the second image segmentation result of the image data to be segmented. The preset text prompt information is the prompt information of the object to which the target segmentation region belongs, and the second image segmentation result is used to characterize the segmentation region where the object to which the target segmentation region belongs is located in the image data to be segmented.

[0086] The target image segmentation result determination module 340 is used to determine the similarity data between the first image segmentation result and the second image segmentation result, and to determine the target image segmentation result of the image data to be segmented based on the similarity data.

[0087] The technical solution of this embodiment involves obtaining image data to be segmented through an initial segmentation result determination module, segmenting the image data to be segmented based on preset channel threshold data to obtain an initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region; a first image segmentation result determination module is used to determine the prompt point information corresponding to each initial segmentation region, inputting the prompt point information and the image data to be segmented into a preset image segmentation model for processing to obtain a first image segmentation result of the image data to be segmented, wherein the first image segmentation result is used to characterize the target segmentation region in the image data to be segmented; a second image segmentation result determination module is used to obtain preset text prompt information, inputting the preset text prompt information and the image data to be segmented into a preset image segmentation model for processing to obtain a second image segmentation result of the image data to be segmented, wherein the preset text prompt information is the prompt information of the object to which the target segmentation region belongs, and the second image segmentation result is used to characterize the segmentation region where the object to which the target segmentation region belongs is located in the image data to be segmented; a target image segmentation result determination module is used to determine the similarity data between the first image segmentation result and the second image segmentation result, and determine the target image segmentation result of the image data to be segmented based on the similarity data. This approach solves the problem of requiring a large amount of labeled data to determine image segmentation, reduces the dependence on large amounts of labeled data, eliminates the need for additional model training, and employs a prompting strategy to enhance the image segmentation model's ability to recognize images in complex environments. This improves the speed, accuracy, and applicability of segmentation processing of target regions in the image data to be segmented.

[0088] Based on the above embodiments, optionally, the preset channel threshold data is preset tone threshold data; the initial segmentation result determination module 310 is specifically used to perform color space conversion processing on the image data to be segmented to obtain image data in a preset color space, the image data in the preset color space including tone channel data; compare the tone channel data of the image data to be segmented with the preset tone threshold data, segment out the region that meets the preset tone threshold data, and obtain the initial segmentation result.

[0089] Optionally, the apparatus further includes an image preprocessing module. Before determining the cue point information corresponding to each initial segmentation region, the image preprocessing module performs image preprocessing on the initial segmentation result to obtain the processed initial segmentation result. The image preprocessing includes morphological processing and filtering processing, and the initial segmentation result includes at least one connected component.

[0090] Optionally, the first image segmentation result determination module 320 is used to traverse each initial segmentation region, input the prompt point information and the image data to be segmented of each initial segmentation region into a preset image segmentation model for processing, and obtain the first sub-segmentation result corresponding to each initial segmentation region. The first sub-segmentation results form the first image segmentation result.

[0091] Optionally, the target image segmentation result determination module 340 is also used to determine the similarity data between the first image segmentation result and the second image segmentation result;

[0092] The similarity data is compared with a preset similarity threshold. If the similarity data is less than the preset similarity threshold, the segmentation result of the first image is used as the segmentation result of the target image.

[0093] Optionally, the first image segmentation result includes at least one first sub-segmentation result corresponding to at least one target segmentation region, and the second image segmentation result includes at least one second sub-segmentation result corresponding to at least one object segmentation region; the target image segmentation result determination module is further configured to traverse the first image segmentation result and the second image segmentation result to determine the similarity data between any first sub-segmentation result and any second sub-segmentation result.

[0094] Optionally, the target image segmentation result determination module 340 is further configured to: if any similarity data is greater than or equal to a preset similarity threshold, determine the target segmentation region corresponding to the similarity data that is greater than or equal to the preset similarity threshold, adjust the prompt point information of the initial segmentation result corresponding to the target segmentation region, and return to execute the step of inputting the prompt point information and the image data to be segmented into a preset image segmentation model for processing based on the adjusted prompt point information, so as to obtain the first image segmentation result of the image data to be segmented, thereby obtaining the target image segmentation result.

[0095] The image segmentation apparatus provided in the embodiments of the present invention can execute the image segmentation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0096] Example 4

[0097] Figure 4 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0098] like Figure 4As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0099] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0100] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as image segmentation methods.

[0101] In some embodiments, the image segmentation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the image segmentation method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the image segmentation method by any other suitable means (e.g., by means of firmware).

[0102] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0103] Computer programs for implementing the image segmentation method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0104] Example 5

[0105] Embodiment 5 of the present invention also provides a computer-readable storage medium storing computer instructions for causing a processor to execute an image segmentation method, the method comprising:

[0106] Acquire image data to be segmented, and perform segmentation processing on the image data to be segmented based on preset channel threshold data to obtain an initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region;

[0107] Determine the cue point information corresponding to each initial segmentation region, input the cue point information and the image data to be segmented into the preset image segmentation model for processing, and obtain the first image segmentation result of the image data to be segmented. The first image segmentation result is used to characterize the target segmentation region in the image data to be segmented.

[0108] Obtain preset text prompt information, input the preset text prompt information and the image data to be segmented into the preset image segmentation model for processing, and obtain the second image segmentation result of the image data to be segmented. The preset text prompt information is the prompt information of the object to which the target segmentation region belongs, and the second image segmentation result is used to characterize the segmentation region where the object to which the target segmentation region belongs is located in the image data to be segmented.

[0109] The similarity data between the first image segmentation result and the second image segmentation result is determined, and the target image segmentation result of the image data to be segmented is determined based on the similarity data.

[0110] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0111] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0112] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0113] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0114] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0115] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An image segmentation method, characterized in that, include: Acquire image data to be segmented, and perform segmentation processing on the image data to be segmented based on preset channel threshold data to obtain an initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region; Determine the cue point information corresponding to each initial segmentation region, input the cue point information and the image data to be segmented into a preset image segmentation model for processing, and obtain a first image segmentation result of the image data to be segmented. The first image segmentation result is used to characterize the target segmentation region in the image data to be segmented. A preset text prompt is obtained, and the preset text prompt and the image data to be segmented are input into a preset image segmentation model for processing to obtain a second image segmentation result of the image data to be segmented. The preset text prompt is the prompt information of the object to which the target segmentation region belongs, and the second image segmentation result is used to characterize the segmentation region in the image data to be segmented where the object to which the target segmentation region belongs is located. The similarity data between the first image segmentation result and the second image segmentation result is determined, and the target image segmentation result of the image data to be segmented is determined based on the similarity data.

2. The method according to claim 1, characterized in that, The preset channel threshold data is preset hue threshold data; The segmentation process based on preset channel threshold data to obtain an initial segmentation result includes: The image data to be segmented is subjected to color space conversion processing to obtain image data in a preset color space, wherein the image data in the preset color space includes tone channel data. The tonal channel data of the image data to be segmented is compared with the preset tonal threshold data to segment out the regions that meet the preset tonal threshold data, thus obtaining the initial segmentation result.

3. The method according to claim 1, characterized in that, Before determining the cue point information corresponding to each of the initial segmented regions, the process includes: The initial segmentation result is subjected to image preprocessing to obtain a processed initial segmentation result, wherein the image preprocessing includes morphological processing and filtering processing, and the initial segmentation result includes at least one connected component.

4. The method according to claim 1, characterized in that, The step of inputting the prompt point information and the image data to be segmented into a preset image segmentation model for processing to obtain a first image segmentation result of the image data to be segmented includes: Each initial segmentation region is traversed, and the prompt point information of each initial segmentation region and the image data to be segmented are input into the preset image segmentation model for processing to obtain the first sub-segmentation result corresponding to each initial segmentation region. Each first sub-segmentation result forms the first image segmentation result.

5. The method according to claim 1, characterized in that, The step of determining the similarity data between the first image segmentation result and the second image segmentation result, and determining the target image segmentation result of the image data to be segmented based on the similarity data, includes: Determine the similarity data between the first image segmentation result and the second image segmentation result; The similarity data is compared with a preset similarity threshold. If the similarity data is less than the preset similarity threshold, the first image segmentation result is used as the target image segmentation result.

6. The method according to claim 5, characterized in that, The first image segmentation result includes at least one first sub-segmentation result corresponding to the target segmentation region, and the second image segmentation result includes at least one second sub-segmentation result corresponding to the segmentation region where the object is located; Determining the similarity data between the first image segmentation result and the second image segmentation result includes: traversing the first image segmentation result and the second image segmentation result to determine the similarity data between any first sub-segmentation result and any second sub-segmentation result.

7. The method according to claim 6, characterized in that, The step of determining the similarity data between the first image segmentation result and the second image segmentation result, and determining the target image segmentation result of the image data to be segmented based on the similarity data, further includes: If any of the similarity data is greater than or equal to the preset similarity threshold, then the target segmentation region corresponding to the similarity data that is greater than or equal to the preset similarity threshold is determined, the prompt point information of the initial segmentation result corresponding to the target segmentation region is adjusted, and based on the adjusted prompt point information, the step of inputting the prompt point information and the image data to be segmented into the preset image segmentation model for processing to obtain the first image segmentation result of the image data to be segmented is returned to execute, so as to obtain the target image segmentation result.

8. An image segmentation apparatus, characterized in that, include: The initial segmentation result determination module is used to acquire image data to be segmented, and to segment the image data to be segmented based on preset channel threshold data to obtain an initial segmentation result, wherein the initial segmentation result includes at least one initial segmentation region. The first image segmentation result determination module is used to determine the prompt point information corresponding to each initial segmentation region, input the prompt point information and the image data to be segmented into a preset image segmentation model for processing, and obtain the first image segmentation result of the image data to be segmented. The first image segmentation result is used to characterize the target segmentation region in the image data to be segmented. The second image segmentation result determination module is used to obtain preset text prompt information, input the preset text prompt information and the image data to be segmented into a preset image segmentation model for processing, and obtain the second image segmentation result of the image data to be segmented. The preset text prompt information is the prompt information of the object to which the target segmentation region belongs, and the second image segmentation result is used to characterize the segmentation region in the image data to be segmented where the object to which the target segmentation region belongs is located. The target image segmentation result determination module is used to determine the similarity data between the first image segmentation result and the second image segmentation result, and to determine the target image segmentation result of the image data to be segmented based on the similarity data.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the image segmentation method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the image segmentation method according to any one of claims 1-7.

Images