Low-efficiency residential land identification method based on semi-supervised generative adversarial network

Abstract

The invention discloses a low-efficiency residential land identification method and device based on a semi-supervised generative adversarial network, a storage medium and electronic equipment. The method comprises the steps of determining a recognition object; based on the identification object, determining a corresponding index, and constructing an index system; obtaining an unlabeled data set, and constructing a labeled data set based on the index system; and training the generative adversarial model based on the unlabeled data set and the labeled data set to obtain a trained generative adversarial model, and inputting a to-be-detected sample into the trained generative adversarial model to obtain a low-efficiency land use identification result. According to the method, a comprehensive index system is established, the deep learning algorithm is introduced into low-efficiency residential land identification, the influence of subjective factors in identification is reduced, the low-efficiency residential land is accurately identified, and the field investigation cost is saved.

Description

Technical Field

[0001] The present application relates to the field of national land space planning, and in particular to a method, device, storage medium and electronic device for identifying inefficient residential land based on a semi-supervised generative adversarial network. Background Art

[0002] Identifying low-efficiency residential land will help develop existing residential land. Against the backdrop of a reduction in new construction land, redevelopment of low-efficiency residential land will be carried out to meet the growing demand for residential land and improve the efficiency of residential land use.

[0003] Currently, identifying inefficient residential land faces several difficulties. First, there is a lack of quantitative standards to measure land use efficiency. The concept of inefficient residential land is relatively abstract, and land use is a complex issue involving many aspects, making it difficult to determine with just one or a few indicators. Second, validating the identification of inefficient residential land typically requires field surveys and other methods to produce convincing results, which consumes significant human and material resources. Finally, existing research on inefficient urban land typically categorizes it into industrial, commercial, and residential land based on its function, constructing indicator systems for identification and evaluation. These indicator systems for identifying and evaluating inefficient residential land are often incomplete, focusing primarily on indicators such as floor area ratio, building density, population density, and land price realization, considering only the functions and benefits of residential land itself. Alternatively, qualitative indicators such as building structure, infrastructure completeness, and planning compliance are considered, but data acquisition is challenging and indicator values are subjective. Weightings are assigned to different indicators based on the indicator system, and this process often involves subjectivity. Summary of the Invention

[0004] The embodiments of the present application provide a method, device, storage medium and electronic device for identifying inefficient residential land based on a semi-supervised generative adversarial network, establish a comprehensive indicator system, introduce a deep learning algorithm into the identification of inefficient residential land, reduce the influence of subjective factors in identification, accurately identify inefficient residential land, and save field investigation costs.

[0005] The present application provides a method for identifying inefficient residential land based on a semi-supervised generative adversarial network, including: Determine the object of identification; Based on the identified objects, corresponding indicators are determined and an indicator system is constructed; Obtain an unlabeled dataset and construct a labeled dataset based on the indicator system; A generative adversarial model is trained based on the unlabeled dataset and the labeled dataset to obtain a trained generative adversarial model, and a sample to be detected is input into the trained generative adversarial model to obtain a low-efficiency land recognition result.

[0006] Furthermore, in the above-mentioned method for identifying inefficient residential land based on a semi-supervised generative adversarial network, the identification object is a cell vector, and the index system includes primary indicators and secondary indicators; The first-level indicators include land use status, economic status, ecological level and living standards; the second-level indicators included in the land use status are: building age, building density, floor area ratio and chaos; the second-level indicators included in the economic status are: night lighting range, commercial focus and average residential price; the second-level indicators included in the ecological level are: park green space status, greening rate and carbon emission content; the second-level indicators included in the living standards are: population density, public services and transportation accessibility.

[0007] Furthermore, in the above-mentioned method for identifying inefficient residential land based on a semi-supervised generative adversarial network, the step of constructing a labeled dataset based on the indicator system includes: Calculate the confidence level of the land use efficiency of the inefficient residential land, and based on the confidence level, label the vector data in the unlabeled dataset as clearly belonging to the category of inefficient residential land and the category of clearly belonging to non-inefficient residential land.

[0008] Furthermore, in the above-mentioned method for identifying inefficient residential land based on a semi-supervised generative adversarial network, the categories of inefficient residential land include old communities, urban villages and other inefficient residential land.

[0009] Furthermore, in the above-mentioned method for identifying inefficient residential land based on a semi-supervised generative adversarial network, the categories of non-inefficient residential land include high-end communities, villas and other non-inefficient residential land.

[0010] Furthermore, in the above-mentioned method for identifying inefficient residential land based on a semi-supervised generative adversarial network, the generative adversarial network includes a generator and a discriminator; The generator is used to receive random noise and generate pseudo samples based on the random noise; The discriminator is used to identify the pseudo samples and the samples to be detected, and obtain a low-efficiency identification result.

[0011] Furthermore, in the above-mentioned method for identifying inefficient residential land based on a semi-supervised generative adversarial network, the step of training the generative adversarial model based on the labeled dataset includes: Sampling a batch of labeled samples from the labeled dataset, sampling a batch of unlabeled samples from the unlabeled dataset, and sampling a batch of noise data from a random distribution; Input the noise data into the generator to obtain a pseudo sample; Inputting the pseudo sample, the labeled sample, and the unlabeled sample into a discriminator to obtain a true sample probability or a false sample probability for each sample, as well as a category probability for the labeled sample; Calculating a loss value of a discriminator based on the true sample probability, the false sample probability, the category probability, and the true label, and updating the parameters of the discriminator through a back-propagation algorithm; Calculate the loss value of the generator based on the pseudo sample and the corresponding false sample probability, and update the parameters of the discriminator through the back propagation algorithm; Repeat the above steps until the generative adversarial model converges.

[0012] The present application also provides a device for identifying inefficient residential land based on a semi-supervised generative adversarial network, including: An identification object determination module, used to determine the identification object; An indicator system construction module, used to determine corresponding indicators based on the identified objects and construct an indicator system; A data set acquisition and construction module is used to acquire an unlabeled data set and construct a labeled data set based on the indicator system; The recognition module is used to train the generative adversarial model based on the unlabeled data set and the labeled data set to obtain a trained generative adversarial model, input the sample to be detected into the trained generative adversarial model, and obtain a low-efficiency land recognition result.

[0013] An embodiment of the present application also provides a computer-readable storage medium, which stores multiple instructions, and the instructions are suitable for being loaded by a processor to execute any of the above-mentioned methods for identifying inefficient residential land based on a semi-supervised generative adversarial network.

[0014] An embodiment of the present application also provides an electronic device, including a processor and a memory, wherein the processor is electrically connected to the memory, the memory is used to store instructions and data, and the processor is used for the steps in any of the above-mentioned methods for identifying inefficient residential land based on a semi-supervised generative adversarial network.

[0015] The present application provides a method, device, storage medium, and electronic device for identifying inefficient residential land based on a semi-supervised generative adversarial network. By introducing a deep learning algorithm into the identification of inefficient residential land, the present application can further explore the relationship between various characteristics of residential land and its land use efficiency based on the deep learning algorithm, reduce the influence of subjective factors in the identification criteria, and make the identification results more accurate and objective. In addition, the model adopts a semi-supervised learning method: combining supervised learning with unsupervised learning, in the case of a small amount of labeled data, using a large amount of unlabeled data for unsupervised learning to improve the performance of supervised learning, and can make full use of a large amount of unlabeled data and reduce dependence on labeled data. In the identification of inefficient residential land, the acquisition of labeled data is often time-consuming and labor-intensive. The use of this method greatly improves the efficiency of recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The following detailed description of the specific embodiments of the present application in conjunction with the accompanying drawings will make the technical solutions and other beneficial effects of the present application apparent.

[0017] Figure 1 A flowchart of a method for identifying inefficient residential land based on a semi-supervised generative adversarial network provided in an embodiment of the present application.

[0018] Figure 2 Another flowchart of the method for identifying inefficient residential land based on a semi-supervised generative adversarial network provided in an embodiment of the present application.

[0019] Figure 3 Schematic diagram of the indicator system provided in the embodiment of this application.

[0020] Figure 4 A schematic diagram of the categories of inefficient residential land and non-inefficient residential land provided in an embodiment of the present application.

[0021] Figure 5 A schematic diagram of the structure of a generative adversarial network provided in an embodiment of the present application.

[0022] Figure 6 A schematic diagram of the structure of an inefficient residential land identification device based on a semi-supervised generative adversarial network provided in an embodiment of the present application.

[0023] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0025] The embodiments of the present application provide a method, device, storage medium, and electronic device for identifying inefficient residential land based on a semi-supervised generative adversarial network. The embodiments of the present application provide a device for identifying inefficient residential land based on a semi-supervised generative adversarial network, which can be integrated into an electronic device, such as a terminal or server. The terminal can include a tablet computer, a laptop computer, a personal computer (PC), a microprocessor box, or other devices.

[0026] See also Figure 1 and Figure 2 , Figure 1 This is a flowchart of a method for identifying inefficient residential land based on a semi-supervised generative adversarial network according to an embodiment of the present application. Figure 2 Another flowchart of a method for identifying inefficient residential land based on a semi-supervised generative adversarial network provided in an embodiment of the present application, which is applied to an electronic device, includes the following steps: S1, determine the identification object.

[0027] Specifically, the identification objects and their characteristics are clearly identified. The identification objects are mainly construction land plots used for residence, which can be plots with blocks as boundaries. In this embodiment, in order to identify more accurately, the plots are further refined into specific cells, and their spatial characteristics are abstracted in their center points.

[0028] S2, based on the identified objects, determines the corresponding indicators and builds an indicator system.

[0029] First, the purpose of identification should be clear. For residential land, the main function is to provide living space for residents. Therefore, the purpose of identification is to screen out land that cannot provide good living space for residents in the same space. Therefore, the identification standard is based on the comfort level of residents and the utilization of space.

[0030] Next, we reviewed relevant literature and theories to analyze the characteristics of the identification object and the existing research indicator systems. We summarized the advantages and disadvantages of different indicator systems and learned from them. The identification object is the cell space vector.

[0031] In one embodiment, the index system includes primary and secondary indicators. The primary indicators include land use status, economic status, ecological level, and living standards; the secondary indicators included in land use status are: building age, building density, volume ratio, and disorder; the secondary indicators included in economic status are: night lighting range, commercial focus, and average residential price; the secondary indicators included in ecological level are: park and green space status, greening rate, and carbon emission content; and the secondary indicators included in living standards are: population density, public services, and transportation accessibility.

[0032] Specifically, according to the specific identification needs, corresponding indicators are selected. The indicators should be able to reflect certain characteristics of residential land. According to the relationship between the indicators, they are summarized into an indicator system. Figure 3 The schematic diagram of the index system provided for the embodiment of the present application is based on four aspects: land use status, economic status, ecological level and living standard. Among them, the land use status mainly involves the building conditions in the land, including building age, building density, volume ratio and disorder. The building age reflects the building quality to a certain extent, the building density and volume ratio reflect the utilization efficiency of the building space, and the disorder reflects the neatness of the building layout. The economic status mainly reflects the economic vitality of the land, which is reflected by the night lighting range, commercial gathering situation and average residential price. Among them, the night lighting range is closely related to the economy, the commercial gathering situation reflects the concentration of commercial facilities in the land area, and the average residential price reflects the overall level of the economy around the land. The ecological level mainly considers the green environmental protection benefits of the land, including the park green space situation, greening rate and carbon emissions. Among them, park green space is a place for residents to relax in their daily lives. Having a large area of park green space in the surrounding area is also a feature that efficient residential land should have. In terms of living standards, it mainly involves the convenience of residents' daily life and the coordination of urban functional layout, including population density, public services and transportation accessibility. Among them, public services and transportation accessibility mainly involve the distribution of public service facilities such as hospitals, schools and roads.

[0033] S3, obtains unlabeled datasets and constructs labeled datasets based on the indicator system.

[0034] The construction of unlabeled datasets is primarily based on an indicator system, using quantitative data to represent them as specific feature datasets. In this example, some data can be directly collected from relevant raster datasets, such as nighttime light data and population density. Some feature data is obtained by web crawling, such as average residential price and building age. Some features are calculated from relevant datasets, such as building density and floor area ratio, which are calculated from building outline vector data.

[0035] In one embodiment, the confidence level of the land use efficiency of inefficient residential land is calculated, and based on the confidence level, vector data in the unlabeled dataset is labeled as clearly belonging to the category of inefficient residential land and clearly belonging to the category of non-inefficient residential land.

[0036] Specifically, the labeled data set is mainly based on the unlabeled data set. According to comprehensive judgment, the judgment results with higher confidence are labeled. Field investigation and other methods can be used. Due to the use of semi-supervised learning methods, the labeled data only accounts for a small part of the entire data set, which can save a lot of manpower and material resources. Figure 4 A schematic diagram of the categories of low-efficiency residential land and non-low-efficiency residential land provided in the embodiment of this application, such as Figure 4 As shown, the categories of low-efficiency residential land include old communities, urban villages and other low-efficiency residential land, and the categories of non-low-efficiency residential land include high-end communities, villas and other non-low-efficiency residential land.

[0037] S4, train the generative adversarial model based on the unlabeled dataset and the labeled dataset to obtain a trained generative adversarial model, input the sample to be tested into the trained generative adversarial model, and obtain the low-efficiency land recognition result.

[0038] Figure 5 A schematic diagram of the structure of the generative adversarial network provided in the embodiment of the present application is shown in FIG. Figure 5 As shown in the figure, the generative adversarial network includes a generator and a discriminator; the generator is used to receive random noise and generate pseudo samples based on the random noise; the discriminator is used to identify pseudo samples and samples to be detected to obtain low-efficiency recognition results.

[0039] Specifically, first, the data set is divided. The annotated data set is divided into a training set, a validation set, and a test set. The training set is the data set used to train the model. The model learns from the training set and adjusts its parameters to minimize the loss function, thereby improving the performance of the model. The validation set is used for model tuning and hyperparameter selection. During the training process, after the model is trained on the training set, it will be evaluated on the validation set to prevent the model from overfitting. The test set is used to finally evaluate the performance of the model. After the model training and validation are completed, the test set is used to evaluate the generalization ability of the model, that is, the performance of the model on completely unknown data. In this embodiment, the data set division ratio is 70% training set, 15% validation set, and 15% test set. In addition, considering that the amount of data in the annotated data set is small, cross-validation is used to make more effective use of the data. Make full use of the data.

[0040] Secondly, the generative adversarial network model based on semi-supervised learning constructed in this embodiment mainly includes two modules: a discriminator module and a generator module. Among them, the discriminator module contains three inputs: labeled samples, unlabeled samples, and pseudo samples generated by the generator. Its main tasks include classification tasks, which only focus on labeled data and divide the labeled data into the correct categories through the softmax function, and discrimination tasks, which distinguish samples generated by the generator from other real samples. In this embodiment, the cross-entropy loss is designed to describe the classification task loss, and the binary cross-entropy loss is designed to describe the discrimination task loss. The loss of the discriminator module is the weighted sum of the above two losses, where the weight is a hyperparameter used to balance the supervised and unsupervised parts and is dynamically adjusted according to the specific situation. The input of the generator is random noise, and its task is to maximize the probability that the generated samples are judged as real samples. Using Wasserstein adversarial loss to train the generator can make the sample distribution generated by the generator closer to the real data distribution and solve the problem of instability in traditional GAN training.

[0041] Next, the model is trained. During training based on the dataset, the generator and discriminator compete against each other. The generator attempts to minimize the difference between the generated data and the real data, making it impossible for the discriminator to distinguish them. The discriminator, on the other hand, attempts to maximize the probability of correctly distinguishing between the real and generated data. First, the generator and discriminator parameters are initialized, followed by iterative training. The specific training steps are as follows: Sample a batch of labeled samples from the labeled dataset, sample a batch of unlabeled samples from the unlabeled dataset, and sample a batch of noise data from the random distribution; Input the noise data into the generator to obtain pseudo samples; Input pseudo samples, labeled samples, and unlabeled samples into the discriminator to obtain the probability of each sample being a true sample or a fake sample, as well as the category probability for the labeled samples; Based on the true sample probability, false sample probability, category probability and true label, the discriminator loss value is calculated, and the discriminator parameters are updated through the back-propagation algorithm; Calculate the loss value of the generator based on the pseudo samples and the corresponding false sample probabilities, and update the parameters of the discriminator through the back-propagation algorithm; Repeat the above steps until the generative adversarial model converges.

[0042] Optimize and adjust the model through data enhancement, hyperparameter optimization and regularization.

[0043] In other embodiments of the present invention, for example, the loss function can be adjusted and designed according to actual needs, gradient penalty can be added, etc., to improve the classification and recognition ability of the model. Finally, there is model testing. In the embodiment, the data in the unlabeled dataset is distinguished using a model with excellent performance tested on the test set to complete the identification of inefficient residential land.

[0044] After the model is trained, it only needs to collect relevant feature data from the target plot's indicator system to identify the current land use efficiency. Based on specific circumstances and changing needs, the dataset training model can be adjusted to adapt to other environments, improving the versatility and efficiency of the identification method while reducing the influence of subjective factors in determining indicator weights in existing methods. Furthermore, unlike current methods that score all identified objects within a unified framework, this method considers multiple categories when annotating data, making it applicable to more complex situations and accurately identifying inefficient use caused by various factors.

[0045] According to the method described in the above embodiment, this embodiment will be further described from the perspective of the identification device of inefficient residential land based on a semi-supervised generative adversarial network. The identification device of inefficient residential land based on a semi-supervised generative adversarial network can be implemented as an independent entity or integrated into an electronic device, which can be a terminal, server and other devices, wherein the terminal can include a tablet computer, a laptop computer, a personal computer (PC, Personal Computer), a micro processing box, or other devices.

[0046] See also Figure 6 , Figure 6 The present invention specifically describes an apparatus for identifying inefficient residential land based on a semi-supervised generative adversarial network, which is applied to an electronic device. The apparatus for identifying inefficient residential land based on a semi-supervised generative adversarial network may include: An identification object determination module, used to determine the identification object; An indicator system construction module, used to determine corresponding indicators based on the identified objects and construct an indicator system; A data set acquisition and construction module is used to acquire an unlabeled data set and construct a labeled data set based on the indicator system; The recognition module is used to train the generative adversarial model based on the unlabeled data set and the labeled data set to obtain a trained generative adversarial model, input the sample to be detected into the trained generative adversarial model, and obtain a low-efficiency land recognition result.

[0047] During specific implementation, the above modules and / or units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above modules and / or units can refer to the previous method embodiments. The specific beneficial effects that can be achieved can also be found in the beneficial effects in the previous method embodiments, which will not be repeated here.

[0048] In addition, an embodiment of the present application further provides an electronic device, which may be a computer, tablet computer, or other device. This electronic device can implement the steps of any embodiment of the method for identifying inefficient residential land based on a semi-supervised generative adversarial network provided in the embodiment of the present application, and thus can achieve the beneficial effects that can be achieved by any method for identifying inefficient residential land based on a semi-supervised generative adversarial network provided in the embodiment of the present application. For details, please refer to the previous embodiment and will not be repeated here.

[0049] Figure 7 The following figure shows a block diagram of the specific structure of an electronic device provided by an embodiment of the present invention. This electronic device can be used to implement the method for identifying inefficient residential land based on a semi-supervised generative adversarial network provided in the above embodiments. The electronic device 500 can be a terminal, server, or other device. The terminal can include a tablet computer, laptop computer, personal computer (PC), microprocessor box, or other device.

[0050] RF circuit 510 is used to receive and transmit electromagnetic waves, converting them into electrical signals, thereby enabling communication with a communications network or other devices. RF circuit 510 may include various existing circuit components for performing these functions, such as an antenna, a radio frequency transceiver, a digital signal processor, an encryption / decryption chip, a subscriber identity module (SIM) card, memory, and the like. RF circuit 510 can communicate with various networks, such as the Internet, an intranet, or a wireless network, or with other devices via a wireless network. These wireless networks may include cellular telephone networks, wireless local area networks, or metropolitan area networks. The wireless networks may utilize various communication standards, protocols, and technologies, including but not limited to Global System for Mobile Communication (GSM), Enhanced Data GSM Environment (EDGE), Wideband Code Division Multiple Access (WCDMA), Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Wireless Fidelity (Wi-Fi) (such as Institute of Electrical and Electronics Engineers standards IEEE 802.11a, IEEE 802.11b, IEEE802.11g, and / or IEEE802.11n), Voice over Internet Protocol (VoIP), Worldwide Interoperability for Microwave Access (Wi-Max), other protocols for email, instant messaging, and short messaging, and any other suitable communication protocols, including those currently undeveloped.

[0051] The memory 520 can be used to store software programs and modules, such as the corresponding program instructions / modules in the above-mentioned embodiments. The processor 580 executes various functional applications and data processing by running the software programs and modules stored in the memory 520, that is, realizing functions such as taking pictures with the front camera, processing the captured images, and switching the display color of the displayed content on the display screen. The memory 520 may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 520 may further include a memory remotely located relative to the processor 580, and these remote memories may be connected to the electronic device 500 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0052] The input unit 530 may be used to receive input digital or character information, and generate a keyboard and a mouse related to user settings and function control. The display unit 540 can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces. These graphical user interfaces can be composed of graphics, text, icons, videos, or any combination thereof. The display unit 540 may include a display panel 541. Optionally, the display panel 541 can be configured in the form of an LCD (Liquid Crystal Display), an OLED (Organic Light-Emitting Diode), or the like.

[0053] Audio circuit 560, speaker 561, and microphone 562 provide an audio interface between the user and electronic device 500. Audio circuit 560 converts received audio data into electrical signals and transmits them to speaker 561, which then converts them into sound signals for output. Microphone 562, on the other hand, converts collected sound signals into electrical signals, which are then received by audio circuit 560 and converted into audio data. The audio data is then processed by output processor 580 and transmitted via RF circuit 510 to, for example, another terminal. Alternatively, the audio data may be output to memory 520 for further processing. Audio circuit 560 may also include an earphone jack to allow communication between external headphones and electronic device 500.

[0054] Electronic device 500, through a transmission module 570 (e.g., a Wi-Fi module), can help users receive requests, send information, and so on, providing users with wireless broadband Internet access. Although the figure shows transmission module 570, it is understood that it is not a required component of electronic device 500 and can be omitted as needed without changing the essence of the invention.

[0055] Processor 580 is the control center of electronic device 500. It connects all components of the phone using various interfaces and circuits. By running or executing software programs and / or modules stored in memory 520 and accessing data stored in memory 520, it executes various functions of electronic device 500 and processes data, thereby providing overall monitoring of the electronic device. Optionally, processor 580 may include one or more processing cores. In some embodiments, processor 580 may integrate an application processor and a modem processor. The application processor primarily handles the operating system, user interface, and application programs, while the modem processor primarily handles wireless communications. It is understood that the modem processor may not be integrated into processor 580.

[0056] Electronic device 500 also includes a power supply 590 (e.g., a battery) for powering various components. In some embodiments, the power supply can be logically connected to processor 580 via a power management system, thereby enabling the power management system to manage charging, discharging, and power consumption. Power supply 590 can also include any components, such as one or more DC or AC power supplies, a recharging system, a power failure detection circuit, a power converter or inverter, and a power status indicator.

[0057] Although not shown, the electronic device 500 also includes a camera (such as a front camera and a rear camera), a Bluetooth module, etc., which will not be described in detail here. Specifically, in this embodiment, the display unit of the electronic device is a touch screen display, and the mobile terminal also includes a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors. The one or more programs include instructions for performing the following operations: During specific implementation, the above modules can be implemented as independent entities, or can be arbitrarily combined and implemented as the same or several entities. The specific implementation of the above modules can be found in the previous method embodiments and will not be repeated here.

[0058] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be accomplished through instructions, or by controlling related hardware through instructions. The instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. To this end, an embodiment of the present invention provides a storage medium storing a plurality of instructions that can be loaded by a processor to execute the steps of any of the embodiments of the method for identifying inefficient residential land based on a semi-supervised generative adversarial network provided in an embodiment of the present invention.

[0059] The computer-readable storage medium may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0060] Since the instructions stored in the storage medium can execute the steps in any embodiment of the method for identifying inefficient residential land based on a semi-supervised generative adversarial network provided in the embodiments of the present invention, the beneficial effects that can be achieved by any method for identifying inefficient residential land based on a semi-supervised generative adversarial network provided in the embodiments of the present invention can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0061] The above is a detailed introduction to the method, device, storage medium and electronic device for identifying inefficient residential land based on a semi-supervised generative adversarial network provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for identifying inefficient residential land based on a semi-supervised generative adversarial network, characterized in that: The method comprises: Determine the object of identification; Based on the identified objects, corresponding indicators are determined and an indicator system is constructed; Obtain an unlabeled dataset and construct a labeled dataset based on the indicator system; A generative adversarial model is trained based on the unlabeled dataset and the labeled dataset to obtain a trained generative adversarial model, and a sample to be detected is input into the trained generative adversarial model to obtain a low-efficiency land recognition result.

2. The method for identifying inefficient residential land based on a semi-supervised generative adversarial network according to claim 1, characterized in that: The identification object is a cell space vector, and the index system includes primary indexes and secondary indexes; The first-level indicators include land use status, economic status, ecological level and living standards; the second-level indicators included in the land use status are: building age, building density, floor area ratio and chaos; the second-level indicators included in the economic status are: night lighting range, commercial focus and average residential price; the second-level indicators included in the ecological level are: park green space status, greening rate and carbon emission content; the second-level indicators included in the living standards are: population density, public services and transportation accessibility.

3. The method for identifying inefficient residential land based on a semi-supervised generative adversarial network according to claim 1, characterized in that: The constructing of the labeled data set based on the indicator system includes: Calculate the confidence level of the land use efficiency of the inefficient residential land, and based on the confidence level, label the vector data in the unlabeled dataset as clearly belonging to the category of inefficient residential land and the category of clearly belonging to non-inefficient residential land.

4. The method for identifying inefficient residential land based on a semi-supervised generative adversarial network according to claim 3, characterized in that: The categories of low-efficiency residential land include old communities, urban villages and other low-efficiency residential land.

5. The method for identifying inefficient residential land based on a semi-supervised generative adversarial network according to claim 4, characterized in that: The categories of non-low-efficiency residential land include high-end residential communities, villas and other non-low-efficiency residential land.

6. The method for identifying inefficient residential land based on a semi-supervised generative adversarial network according to claim 1, characterized in that: The generative adversarial network includes a generator and a discriminator; The generator is used to receive random noise and generate pseudo samples based on the random noise; The discriminator is used to identify the pseudo samples and the samples to be detected, and obtain a low-efficiency identification result.

7. The method for identifying inefficient residential land based on a semi-supervised generative adversarial network according to claim 6, characterized in that: The training of the generative adversarial model based on the labeled data set includes: Sampling a batch of labeled samples from the labeled dataset, sampling a batch of unlabeled samples from the unlabeled dataset, and sampling a batch of noise data from a random distribution; Input the noise data into the generator to obtain a pseudo sample; Inputting the pseudo sample, the labeled sample, and the unlabeled sample into a discriminator to obtain a true sample probability or a false sample probability for each sample, as well as a category probability for the labeled sample; Calculating a loss value of a discriminator based on the true sample probability, the false sample probability, the category probability, and the true label, and updating the parameters of the discriminator through a back-propagation algorithm; Calculate the loss value of the generator based on the pseudo sample and the corresponding pseudo sample probability, and update the parameters of the discriminator through the back propagation algorithm; Repeat the above steps until the generative adversarial model converges.

8. A device for identifying inefficient residential land based on a semi-supervised generative adversarial network, characterized in that: include: An identification object determination module, used to determine the identification object; An indicator system construction module, used to determine corresponding indicators based on the identified objects and construct an indicator system; A data set acquisition and construction module is used to acquire an unlabeled data set and construct a labeled data set based on the indicator system; The recognition module is used to train the generative adversarial model based on the unlabeled data set and the labeled data set to obtain a trained generative adversarial model, input the data to be evaluated into the trained generative adversarial model, and obtain a low-efficiency land recognition result.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a plurality of instructions, which are suitable for being loaded by a processor to execute the method for identifying inefficient residential land based on a semi-supervised generative adversarial network as described in any one of claims 1 to 7.

10. An electronic device, characterized in that: It includes a processor and a memory, the processor is electrically connected to the memory, the memory is used to store instructions and data, and the processor is used to execute the steps in the method for identifying inefficient residential land based on a semi-supervised generative adversarial network as described in any one of claims 1 to 7.

Images