Industrial land intelligent evaluation system and method based on artificial intelligence

Abstract

The invention discloses an intelligent industrial land evaluation system and method based on artificial intelligence, and the system obtains enterprise demands and land resource data in real time through a data collection module, guarantees the timeliness of evaluation information, and effectively avoids information lag. The dynamic decision-making module fuses an improved decision-making tree and a DQN reinforcement learning algorithm, can optimize a decision-making strategy online according to a real-time data stream, adapts to market and policy changes, and solves the problem that a traditional method depends on static data and is difficult to deal with a dynamic scene; the decision tree is improved to improve the adaptability of the model to a complex scene, the DQN algorithm automatically learns an optimal strategy, manual intervention is reduced, subjective deviation is reduced, and decision scientificity and generalization ability are enhanced; the index generation module calculates a land efficiency index and an industrial gathering index, provides a multi-dimensional evaluation basis, breaks through the limitation of a single index, more accurately reflects the economic, environmental and industrial values of the land parcels, and enables the evaluation result to better meet the actual demands.

Description

Technical Field

[0001] This invention relates to the field of industrial land assessment technology, and in particular to an intelligent assessment system and method for industrial land based on artificial intelligence. Background Technology

[0002] With the acceleration of urbanization and the upgrading of industrial structure, the rational planning and efficient use of industrial land has become a key factor in promoting economic development. However, traditional industrial land assessment methods often rely on static data and human experience, which have problems such as information lag, strong decision-making subjectivity, and difficulty in adapting to dynamic market changes. Summary of the Invention

[0003] In view of this, the present invention proposes an intelligent assessment system and method for industrial land based on artificial intelligence, which can effectively solve the defects of existing technologies, such as reliance on static data and human experience leading to information lag, strong decision-making subjectivity, and difficulty in adapting to dynamic market changes.

[0004] The technical solution of this invention is implemented as follows: An intelligent assessment system for industrial land use based on artificial intelligence, comprising: The data acquisition module is used to collect enterprise demand data and land resource data in real time; The dynamic decision-making module is used to generate and optimize strategies based on the improveable decision tree and DQN reinforcement learning algorithm, using real-time collected enterprise demand data and land resource data, and generate the structure and parameters of a multi-level decision tree. The structure and parameters of the multi-level decision tree are used to guide land allocation and decision-making. The index generation module is used to calculate the land use efficiency index and the industrial agglomeration index based on the actual data of the land parcels. The output module is used to comprehensively evaluate industrial land based on land use efficiency index, industrial agglomeration index and information related to land allocation and decision-making, and output the evaluation results.

[0005] As a further optional solution to the aforementioned AI-based intelligent assessment system for industrial land, the system utilizes an improveable decision tree and DQN reinforcement learning algorithm to generate and optimize strategies based on real-time collected enterprise demand data and land resource data, generating a multi-level decision tree structure and parameters. The structure and parameters of this multi-level decision tree are used to guide land allocation and decision-making, specifically including: An initial decision tree model is constructed using an improved decision tree algorithm; Based on real-time collected enterprise demand data and land resource data, the node splitting threshold of the improveable decision tree algorithm is dynamically adjusted to obtain the dynamically adjusted decision tree model. By combining the dynamically adjusted decision tree model with the DQN reinforcement learning algorithm, and continuously optimizing the decision strategy by defining the state space, action space and reward function, a DQN reinforcement learning strategy is obtained. Based on the dynamically adjusted decision tree model and DQN reinforcement learning strategy, a land parcel allocation strategy is dynamically generated.

[0006] As a further optional solution to the aforementioned AI-based intelligent assessment system for industrial land use, the specific calculation formula used for adjusting the node splitting threshold is as follows: Tnew=Told+η×(Dcurrent-Dthreshold); Where Tnew is the updated node splitting threshold, Told is the original node splitting threshold, η is the learning rate, Dcurrent is the real-time data stream feature value, and Dthreshold is the preset threshold.

[0007] As a further optional solution to the aforementioned AI-based intelligent assessment system for industrial land use, the specific formula for calculating the land use efficiency index is as follows: E = (Annual output value of the enterprise / Land area) × W1 + (Tax contribution / Land area) × W2; Among them, the weights W1 and W2 are dynamically optimized using the entropy weight method based on the enterprise's annual output value, tax contribution, and land area data.

[0008] As a further optional solution to the aforementioned AI-based intelligent assessment system for industrial land, the step of comprehensively assessing industrial land based on land use efficiency index, industrial agglomeration index, and information related to land allocation and decision-making, and outputting the assessment results, specifically includes: The land use efficiency index, industrial agglomeration index and land parcel decision information are weighted and integrated to calculate a comprehensive evaluation score; Based on the comprehensive evaluation score, a comprehensive evaluation report for the land parcel is generated. The comprehensive evaluation report includes land use efficiency level, industrial agglomeration degree, land parcel allocation suggestions and optimization directions, and outputs the results through a visualization interface or data interface.

[0009] As a further optional solution to the aforementioned AI-based intelligent assessment system for industrial land use, the land use efficiency index, industrial agglomeration index, and land parcel decision information are weighted and integrated, and the specific calculation formula is as follows: ; Where S is the overall evaluation score. , , Here are the weights for each indicator: E is the land use efficiency index, PAI is the industrial agglomeration index, and D is the quantitative score of land parcel decision-making information.

[0010] An intelligent assessment method for industrial land use based on artificial intelligence, specifically including: Data collection steps: Real-time collection of enterprise demand data and land resource data; Dynamic decision-making steps: Based on the improveable decision tree and DQN reinforcement learning algorithm, strategy generation and optimization are performed using real-time collected enterprise demand data and land resource data to generate the structure and parameters of a multi-level decision tree. The structure and parameters of the multi-level decision tree are used to guide land allocation and decision-making. Index generation steps: Calculate the land use efficiency index and industry agglomeration index based on the actual data of the land parcels; Comprehensive assessment and output steps: Based on the land use efficiency index, industry agglomeration index, and information related to land allocation and decision-making, conduct a comprehensive assessment of industrial land use and output the assessment results.

[0011] A computing device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described intelligent assessment method for industrial land use based on artificial intelligence.

[0012] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described intelligent assessment method for industrial land use based on artificial intelligence.

[0013] The beneficial effects of this invention are as follows: The data acquisition module acquires enterprise demand and land resource data in real time, ensuring the timeliness of basic assessment information. The dynamic decision-making module, combining an improved decision tree (supporting dynamic threshold adjustment) and the DQN reinforcement learning algorithm, can optimize decision-making strategies online based on real-time data streams, adapting to rapid market and policy changes. This solves the information lag problem caused by traditional methods relying on static data, making the assessment results closer to actual needs. The dynamic decision-making module, through an improveable decision tree, enhances the model's adaptability to complex scenarios, overcoming the shortcomings of traditional methods in adapting to dynamic market changes. Simultaneously, the dynamic decision-making module, through the DQN reinforcement learning algorithm, automatically learns the optimal strategy, reducing human experience intervention and mitigating biases caused by subjective rules in traditional methods, thus improving the scientific nature and generalization ability of decision-making. The index generation module calculates the land use efficiency index (integrating multi-dimensional data such as output value, tax revenue, and energy consumption) and the industrial agglomeration index (based on location entropy analysis), providing comprehensive assessment basis and breaking through the limitations of single-indicator assessment, more accurately reflecting the economic, environmental, and industrial value of the land parcel. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the 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.

[0015] Figure 1 This is a schematic diagram illustrating the composition of an intelligent assessment system for industrial land based on artificial intelligence, as described in this invention. Figure 2 This is a flowchart illustrating an intelligent assessment method for industrial land based on artificial intelligence, as described in this invention. Figure 3 This is a schematic diagram of the components of a computing device. Detailed Implementation

[0016] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] refer to Figures 1 to 3 An intelligent assessment system for industrial land use based on artificial intelligence includes a data acquisition module, a dynamic decision-making module, an index generation module, and an output module, wherein: The data acquisition module is used to collect enterprise demand data (such as production capacity, energy consumption, and logistics) and land resource data (such as area, location, and policies) in real time.

[0018] The dynamic decision-making module is used to generate and optimize strategies based on the improveable decision tree and DQN reinforcement learning algorithm, using real-time collected enterprise demand data and land resource data, to generate the structure and parameters of a multi-level decision tree. The structure and parameters of the multi-level decision tree are used to guide land allocation and decision-making.

[0019] In some embodiments, the improved decision tree and DQN reinforcement learning algorithm utilizes real-time collected enterprise demand data and land resource data to generate and optimize strategies, generating a multi-level decision tree structure and parameters. The structure and parameters of the multi-level decision tree are used to guide land allocation and decision-making, specifically including: An improved decision tree algorithm is used to construct an initial decision tree model based on historical data or pre-defined rules; During system operation, based on real-time collected enterprise demand data (covering capacity planning, energy consumption demand, logistics and distribution requirements, etc.) and land resource data (including land area, geographical location, policy constraints, etc.), the node splitting threshold of the improveable decision tree algorithm is dynamically adjusted to obtain the dynamically adjusted decision tree model. By combining the dynamically adjusted decision tree model with the DQN reinforcement learning algorithm, the decision-making strategy is continuously optimized by defining the state space (including land attributes, enterprise demand characteristics, etc.), action space (land allocation / rejection / temporary decision, etc.) and reward function (e.g., R = land use efficiency index improvement value × 0.7 + policy compliance rate × 0.3, which can be adjusted according to the actual situation). The DQN algorithm dynamically evaluates the decision effect based on the real-time feedback reward signal and updates the Q value table, thereby obtaining the DQN reinforcement learning strategy. Based on a dynamically adjusted decision tree model and DQN reinforcement learning strategy, this system comprehensively considers various factors such as enterprise needs, land resource conditions, policy constraints, and market environment to dynamically generate land allocation strategies. These strategies can adaptively adjust according to real-time data changes, providing a scientific and reasonable basis for industrial land allocation.

[0020] Specifically, by collecting real-time enterprise demand data (such as capacity planning, energy consumption demand, and logistics and distribution requirements) and land resource data (such as land area, geographical location, and policy constraints), the system can dynamically reflect changes in the market and resources, avoid decision-making lag caused by traditional static data, and the node splitting threshold of the decision tree algorithm can be dynamically optimized according to real-time data, making the model more flexible to adapt to complex and ever-changing data distribution, and improving the timeliness and accuracy of decision-making. The state space definition encompasses multi-dimensional information such as land attributes and enterprise demand characteristics, enabling the decision-making model to comprehensively consider key factors in land allocation and reduce the limitations of a single perspective. The DQN algorithm continuously optimizes the decision-making strategy through reward functions (such as the improvement value of land use efficiency index and policy compliance rate), dynamically evaluates the decision-making effect and updates parameters, thereby generating a land allocation scheme that is more in line with actual needs. The structure and parameters of the multi-level decision tree model can intuitively display the logical chain of land use decision-making (such as which enterprise demand characteristics or land attributes have a greater impact on the allocation results), enhance the interpretability of the decision, and facilitate the understanding of policymakers and enterprises. The design of the reward function (such as the weight allocation of land use efficiency and policy compliance) quantifies and integrates policy objectives (such as low-carbon development) into the decision-making process, ensuring the scientific nature and policy orientation of the strategy. The system comprehensively considers various factors such as enterprise needs, land resource conditions, policy constraints, and market environment. It dynamically adjusts strategies to cope with complex scenarios, avoiding the shortcomings of traditional methods that are difficult to adapt to changes due to fixed rules. The DQN algorithm continuously updates the strategy through real-time feedback reward signals, enabling the system to adapt to data changes and provide a long-term stable optimization path for industrial land allocation.

[0021] In some embodiments, the specific calculation formula used to adjust the node splitting threshold is as follows: Tnew=Told+η×(Dcurrent-Dthreshold); Where Tnew is the updated node splitting threshold; Told is the original node splitting threshold; η is the learning rate, which controls the step size of parameter updates. The choice of learning rate is crucial to the convergence speed and stability of the algorithm. If the learning rate is too large, it may lead to algorithm instability or missing the optimal solution; if the learning rate is too small, it may lead to slow convergence speed; Dcurrent is the real-time data stream feature value, that is, the data feature value collected at the current moment. This value is used to reflect the current data state or environment; Dthreshold is the preset threshold, which is an initial value set based on historical data or experience, and is used as the benchmark for node splitting.

[0022] Specifically, by introducing the real-time data stream feature value Dcurrent, the system can dynamically capture changes in the current data state or environment, enabling the node splitting threshold T to be adjusted in real time according to the data features. This avoids the model rigidity caused by traditional fixed thresholds. The threshold adjustment formula allows the decision tree model to adapt to the data distribution at different stages. For example, when there are significant changes in enterprise needs or land resource characteristics, the splitting conditions can be automatically optimized to improve the timeliness of decision-making. The learning rate η controls the step size of parameter updates, avoiding algorithm instability or missing the optimal solution (such as oscillation) due to excessively large step size, and also preventing slow convergence due to excessively small step size. By adjusting η, the learning speed and stability of the model can be balanced according to the specific needs of the scenario, such as quickly adapting to data changes in the early stage and finely optimizing the threshold in the later stage. The initial threshold Dthreshold, set based on historical data or experience, provides a stable starting point for the model and avoids the cold start problem. The formula calculates the deviation between real-time data and the benchmark by (Dcurrent − Dthreshold) and dynamically adjusts the threshold, so that the model can both utilize historical experience and adapt to new data trends. Dynamically adjusted Tnew makes the node splitting of the decision tree more closely match the actual data characteristics. For example, in areas with intensive enterprise demand or scarce land resources, the threshold may be automatically tightened to improve the accuracy of land allocation. Compared with a fixed threshold, a dynamic threshold can better balance the misjudgment rate (such as incorrect allocation of land parcels) and improve the reliability of the decision model.

[0023] The index generation module is used to calculate the land use efficiency index and the industrial agglomeration index based on the actual data of the land parcel.

[0024] In some embodiments, the specific formula for calculating the land use efficiency index is as follows: E = (Annual output value of the enterprise / Land area) × W1 + (Tax contribution / Land area) × W2; Among them, the weights W1 and W2 are dynamically optimized using the entropy weight method based on the enterprise's annual output value, tax contribution, and land area data.

[0025] Specifically, the formula uses two core economic indicators—annual output value and tax contribution—combined with land area to comprehensively measure the economic output efficiency of a plot of land. This avoids the one-sidedness of a single indicator (such as using only output value). By dividing output value and tax revenue by land area, the impact of differences in plot size is eliminated, making plots of different sizes comparable and improving the fairness of the assessment. The entropy weight method is used to dynamically determine the weights W1 and W2. The weights are assigned based on the dispersion of the data itself, which reduces human subjective intervention and makes the weights more in line with the actual data characteristics. The weights are automatically adjusted as the data changes. For example, if the data of tax contribution is more dispersed (lower entropy value) in a certain period, its weight W2 will increase accordingly, reflecting that the indicator has a greater impact on land use efficiency in that stage.

[0026] It should be noted that the weights W1 and W2 are dynamically optimized using the entropy weight method based on the company's annual output value, tax contribution, and land area data, specifically including: Data collection and indicator construction: Collect data on the company's annual output value (X1), tax contribution (X2), and land area (X3) to construct a land use efficiency evaluation indicator system; Data standardization: The collected raw data Xi are standardized to eliminate the influence of dimensions. The standardization formula is: X'=(X-μ) / σ, where μ is the mean and σ is the standard deviation. Entropy weight method for dynamic optimization of weights: Based on standardized data, the weights of each indicator are calculated using the entropy weight method. Calculate the weight of each indicator. : ; Calculate the entropy value of each indicator. : ; Calculate the coefficient of variation for each indicator. : ; Determine the weight of each indicator : ; Exponential normalization: The calculated land use efficiency index E is normalized to eliminate differences in magnitude. The normalization formula is as follows: ; in and These are the maximum and minimum values ​​of the land use efficiency index, respectively.

[0027] In addition, the industrial agglomeration index is calculated as follows: A = α × Firm density index + β × Supply chain completeness Where: Enterprise density index = number of enterprises in the target industry / total area of ​​the park (km²); Supply chain completeness = ∑ Matching degree of upstream and downstream enterprises / Total number of links in the target supply chain; Weight settings: α = 0.6 (firm density weight) β=0.4 (weight of industrial chain integrity).

[0028] The output module is used to comprehensively evaluate industrial land based on land use efficiency index, industrial agglomeration index and information related to land allocation and decision-making, and output the evaluation results.

[0029] In some embodiments, the comprehensive assessment of industrial land based on land use efficiency index, industrial agglomeration index, and information related to land allocation and decision-making, and the output of the assessment results, specifically includes: The land use efficiency index, industrial agglomeration index and land parcel decision information are weighted and integrated to calculate a comprehensive evaluation score; Based on the comprehensive evaluation score, a comprehensive evaluation report for the land parcel is generated. The comprehensive evaluation report includes land use efficiency level, industrial agglomeration degree, land parcel allocation suggestions and optimization directions, and outputs the results through a visualization interface or data interface.

[0030] Specifically, by weighting and integrating the land use efficiency index (reflecting economic output efficiency), the industrial agglomeration index (reflecting industrial spatial synergy), and land parcel decision-making information (such as policy compliance and demand matching), the system can comprehensively consider the multi-dimensional value of land parcels and avoid the one-sidedness of single indicator evaluation. The comprehensive evaluation score quantifies the merits and demerits of land parcels in numerical form, making it easy for decision-makers to quickly compare and rank them. For example, resources can be prioritized for high-scoring land parcels. The evaluation report includes land parcel allocation suggestions and optimization directions (such as adjusting industry types and improving land use efficiency), providing practical guidance for actual decision-making. By outputting evaluation results through a visual interface (such as heat maps showing land use efficiency levels and industrial agglomeration distribution), decision-makers can quickly locate problem areas or potential plots, reduce manual analysis costs, and support outputting results through data interfaces, which facilitates integration with other systems (such as urban planning platforms and enterprise service systems) to achieve cross-departmental collaborative decision-making.

[0031] In some embodiments, the weighted fusion of land use efficiency index, industrial agglomeration index, and land parcel decision information is specifically calculated using the following formula: ; Where S is the overall evaluation score. , , Here are the weights for each indicator: E is the land use efficiency index, PAI is the industrial agglomeration index, and D is the quantitative score of land parcel decision-making information.

[0032] An intelligent assessment method for industrial land use based on artificial intelligence, specifically including: Data collection steps: Real-time collection of enterprise demand data and land resource data; Dynamic decision-making steps: Based on the improveable decision tree and DQN reinforcement learning algorithm, strategy generation and optimization are performed using real-time collected enterprise demand data and land resource data to generate the structure and parameters of a multi-level decision tree. The structure and parameters of the multi-level decision tree are used to guide land allocation and decision-making. Index generation steps: Calculate the land use efficiency index and industry agglomeration index based on the actual data of the land parcels; Comprehensive assessment and output steps: Based on the land use efficiency index, industry agglomeration index, and information related to land allocation and decision-making, conduct a comprehensive assessment of industrial land use and output the assessment results.

[0033] A computing device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described intelligent assessment method for industrial land use based on artificial intelligence.

[0034] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described intelligent assessment method for industrial land use based on artificial intelligence.

[0035] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent assessment system for industrial land use based on artificial intelligence, characterized in that, include: The data acquisition module is used to collect enterprise demand data and land resource data in real time; The dynamic decision-making module is used to generate and optimize strategies based on the improveable decision tree and DQN reinforcement learning algorithm, using real-time collected enterprise demand data and land resource data, and to generate the structure and parameters of a multi-level decision tree. The structure and parameters of the multi-level decision tree are used to guide land allocation and decision-making. The index generation module is used to calculate the land use efficiency index and the industrial agglomeration index based on the actual data of the land parcels. The output module is used to comprehensively evaluate industrial land based on land use efficiency index, industrial agglomeration index and information related to land allocation and decision-making, and output the evaluation results.

2. The intelligent assessment system for industrial land based on artificial intelligence according to claim 1, characterized in that, The improved decision tree and DQN reinforcement learning algorithm utilizes real-time collected enterprise demand data and land resource data to generate and optimize strategies, producing a multi-level decision tree structure and parameters. This multi-level decision tree structure and parameters guide land allocation and decision-making, specifically including: An initial decision tree model is constructed using an improved decision tree algorithm; Based on real-time collected enterprise demand data and land resource data, the node splitting threshold of the improveable decision tree algorithm is dynamically adjusted to obtain the dynamically adjusted decision tree model. By combining the dynamically adjusted decision tree model with the DQN reinforcement learning algorithm, and continuously optimizing the decision strategy by defining the state space, action space and reward function, a DQN reinforcement learning strategy is obtained. Based on the dynamically adjusted decision tree model and DQN reinforcement learning strategy, a land parcel allocation strategy is dynamically generated.

3. The intelligent assessment system for industrial land use based on artificial intelligence according to claim 2, characterized in that, The specific calculation formula used to adjust the node splitting threshold is as follows: Tnew=Told+η×(Dcurrent-Dthreshold); Where Tnew is the updated node splitting threshold, Told is the original node splitting threshold, η is the learning rate, Dcurrent is the real-time data stream feature value, and Dthreshold is the preset threshold.

4. The intelligent assessment system for industrial land use based on artificial intelligence according to claim 3, characterized in that, The specific formula for calculating the land use efficiency index is as follows: E = (Annual output value of the enterprise / Land area) × W1 + (Tax contribution / Land area) × W2; Among them, the weights W1 and W2 are dynamically optimized using the entropy weight method based on the enterprise's annual output value, tax contribution, and land area data.

5. The intelligent assessment system for industrial land based on artificial intelligence according to claim 4, characterized in that, The process involves a comprehensive evaluation of industrial land use based on land use efficiency index, industrial agglomeration index, and information related to land allocation and decision-making, with the evaluation results output. Specifically, this includes: The land use efficiency index, industrial agglomeration index and land parcel decision information are weighted and integrated to calculate a comprehensive evaluation score; Based on the comprehensive evaluation score, a comprehensive evaluation report for the land parcel is generated. The comprehensive evaluation report includes land use efficiency level, industrial agglomeration degree, land parcel allocation suggestions and optimization directions, and outputs the results through a visualization interface or data interface.

6. The intelligent assessment system for industrial land based on artificial intelligence according to claim 5, characterized in that, The specific formula for weighted integration of land use efficiency index, industrial agglomeration index and land parcel decision information is as follows: ； Where S is the overall evaluation score. , , Here are the weights for each indicator: E is the land use efficiency index, PAI is the industrial agglomeration index, and D is the quantitative score of land parcel decision-making information.

7. An intelligent assessment method for industrial land use based on artificial intelligence, characterized in that, Specifically, it includes: Data collection steps: Real-time collection of enterprise demand data and land resource data; Dynamic decision-making steps: Based on the improveable decision tree and DQN reinforcement learning algorithm, strategy generation and optimization are performed using real-time collected enterprise demand data and land resource data to generate the structure and parameters of a multi-level decision tree. The structure and parameters of the multi-level decision tree are used to guide land allocation and decision-making. Index generation steps: Calculate the land use efficiency index and industry agglomeration index based on the actual data of the land parcels; Comprehensive assessment and output steps: Based on the land use efficiency index, industry agglomeration index, and information related to land allocation and decision-making, conduct a comprehensive assessment of industrial land use and output the assessment results.

8. A computing device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the AI-based intelligent assessment method for industrial land as described in claim 7.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the intelligent assessment method for industrial land based on artificial intelligence as described in claim 7.

Images