Method for analyzing spatial environment semantics and element association based on empowerment photography

Abstract

The application discloses a space environment semantic and element correlation analysis method based on empowerment photography. The method comprises the following steps: recruiting public participants and carrying out training; the public participants independently take photos in a specific space environment without monitoring, and independently select photos and tell the reasons for taking the photos during the meeting; researchers transcribe the interview audio into text, analyze the syntactic semantic structure into a tree diagram by using the application assembly theory, extract the temporal, syntactic features in the language mode to construct a semantic comparison sequence; meanwhile, the spatial environment element features in the photos are extracted by using a digital image matching technology to construct a space reference sequence; then, a grey correlation analysis model is used to determine the correlation coupling relationship between the semantic of the spatial environment element and the temporal, syntactic and semantic in the language mode, and qualitative graphic data of empowerment photography and photo interview are obtained to represent the overall cognition of the social public to the specific space environment, thereby providing data guidance for space environment reconstruction.

Description

[0001] Parent case information: Patent application number: 2022104693600; Invention title: Application method of empowered photography technology in space environment survey; Application date: 2022-04-30. Technical Field

[0002] This invention belongs to the field of spatial environment survey technology, specifically involving a method for analyzing spatial environment semantics and element associations based on weighted photography. Background Technology

[0003] Traditional spatial environment surveys primarily employ questionnaires and interviews, with analyses often relying on qualitative or statistical quantitative methods. These approaches have limitations in terms of the scientific rigor and objectivity of data acquisition. Questionnaires can only collect public responses to pre-defined questions or simple answers to open-ended questions. While this data has irreplaceable advantages in large-scale quantitative statistics and analysis, the questions themselves are limited by the questionnaire creator's understanding and cannot fully represent the public's overall perception of a specific spatial environment, thus hindering the exploration of deeper underlying causes. Interviews, while capable of uncovering in-depth information, rely primarily on written communication, resulting in weak connections to three-dimensional spatial environment elements and impeding effective information transmission. Furthermore, these methods are researcher-driven and lack objective and detailed analysis based on public perception. Therefore, a more scientific, objective, and concrete spatial environment survey and analysis method based on a "bottom-up" logic is needed. This method should fully reflect the public's overall perception of a specific spatial environment, explore the underlying patterns related to the specific spatial environment and the public's overall perception within that environment, and guide relevant practices.

[0004] Existing survey methods are mostly researcher-led, collecting data through pre-set frameworks, making it difficult to obtain the public's true subjective spatial perceptions, and existing analytical methods are difficult to establish a scientific connection between the public's subjective expressions and physical spatial elements. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a spatial environment semantic and element correlation analysis method based on weighted photography. The aim is to obtain a relatively objective and in-depth understanding of the public's perception of a specific spatial environment, establish a semantic decoding analysis method for interview texts based on the combination theory, and a gray-scale correlation analysis method for spatial environment material elements and semantic elements, thereby improving the objectivity and scientific nature of the public's overall perception of a specific spatial environment.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A method for analyzing the semantics and element associations of spatial environment based on empowered photography involves researchers of specific spatial environments and the general public, with those participating in spatial environment surveys being public participants; the method is carried out in the following steps: Step 1: The public learns about the basic information of the project; the researchers introduce the basic information of the project to the public, including the research team, research topic, research objectives, research significance, project timeline, and project confidentiality matters, etc. Step Two: Public Registration for the Project; Researchers recruit members of the public who are interested in the specific spatial environment and willing to participate in the project; Step 3: Researchers and public participants define their respective rights and obligations; researchers record the basic information of public participants and provide them with photography training, including but not limited to clarifying the research topic, how to use photographic equipment, and determining the time and scope of photography; researchers and public participants sign confidentiality agreements and authorization letters, and agree on a meeting time and place; Step 4: Each public participant takes photographs in a specific spatial environment; Each public participant moves freely in a specific spatial environment and takes photographs independently without being monitored by researchers, using the photos to demonstrate their views on the research topic, as well as their experience and understanding of the research topic related to the specific spatial environment. Step 5: End photography and begin the meeting; All public participants will end their photography session and both parties will meet at the agreed time and place. Each public participant will meet with the researcher separately. Step Six: Selecting and Explaining Photos; During the meeting, public participants select photos they wish to share from the photos taken, and decide the order in which they will share their experiences and understanding of the research topic; Researchers observe and record the public participants' selection process and the photos they choose, and prepare equipment for recording the interview; Public participants explain their photos to the researchers and explain why they took them, with the explanations revolving around the research topic; Researchers listen actively and do not interrupt, comment on, or guide the conversation until the public participants have finished speaking, only responding to the interview when necessary; Step 7: End the interview; After completing Step 6 with one public participant, the researcher will proceed to Step 6 with the next public participant, until the researcher has completed Step 6 with all public participants, thus ending all interviews. Step 8: Organize the data; Researchers organize the information forms, empowered photographs, and interview audio recordings of all public participants, and classify and number the obtained data. Step Nine: Researchers transcribed the interview audio into text, used collocation theory to encode and decode the semantic elements of the interview text, parsed the syntactic and semantic structure of the interview text into a tree diagram, extracted the tense and syntactic features in the language pattern, and constructed a semantic comparison sequence. At the same time, they extracted the spatial environment element features in the photos through digital image matching technology and constructed a spatial reference sequence. Then, they used a gray-scale correlation analysis model to determine the semantic relationship between the semantics of the spatial environment elements and the tense, syntax, and semantics in the language pattern, and obtained qualitative graphic data of weighted photography and photo interviews, which represent the overall public perception of a specific spatial environment.

[0007] In step six above, public participants explain their reasons for taking the photos, establishing a correspondence between the photos and the interview content, thus providing input for subsequent analysis. In step nine, tense and syntactic features of the interview text are extracted using combinatorial theory to construct a semantic comparison sequence, and spatial environmental features in the photos are extracted using digital image matching technology to construct a spatial reference sequence. Finally, the correlation and coupling relationship between the two is determined using a gray-scale correlation analysis model.

[0008] In step six, the interview response under the necessary circumstances includes one of the following four scenarios: Scenario 1: When public participants show a lack of confidence or little interest, researchers can make an attentional response to show their concern and appreciation for the public participants, so as to encourage them to express their true photographic situation and their experience and understanding of the specific spatial environment. Scenario 2: When researchers are unclear about the meaning or attitude of public participants' statements, they can make a clarifying response to confirm the content of the public participants' conversations. Scenario 3: When researchers want to know more details about what public participants are saying, they can make a follow-up questioning response, asking more questions about the content of the public participants' conversations. Scenario 4: When public participants stop talking, researchers can provide a guided continuation response to encourage them to talk more, or researchers can carefully examine the content of the photographs and provide guided continuation responses to the objects in the images that the public participants did not talk about.

[0009] In step nine, the encoding and decoding of semantic elements in the interview text using the application of grouping theory includes the following steps: (i): Parse the syntactic and semantic structure of the interview text into a tree diagram, starting from the top of the tree diagram of the syntactic and semantic structure of the interview text, that is, from the main clause or the main predicate structure; (ii): Find a term contained in the tree diagram of the semantic structure discussed in the photograph; (iii): Perform word, syntactic and semantic analysis using the methods in the next step, i.e., step (iv); (iv): The specific process for applying combination limit encoding and decoding to the entries is as follows: (a) Subtract the semantic or syntactic description from the tree diagram that the researcher started with; (b) Add semantics or syntax to the tree diagram that the researcher is drawing; (c) Repeat steps (ii) and (iii) to continue adding semantic or syntactic descriptions to the tree diagram until the vocabulary in the tree diagram that the researcher started with is used up; (d) Test whether the tree diagram completed by the researchers is qualified.

[0010] In step nine, the adoption of the gray-scale correlation analysis model includes the following steps: (i): The analysis sequence is determined according to the evaluation purpose of the space environment elements, including the reference sequence and the comparison sequence; wherein, the reference sequence refers to the real physical information of the space environment element characteristics shown in the photograph, which is determined by digital image matching technology; the comparison sequence is the features that affect the changes of the space environment elements, which is determined by semantic application combination. (II): Determine the indicator data and reference data columns; (iii): Dimensionless processing of indicator data; (iv): Calculate the absolute difference between the corresponding elements of the indicator sequence and the reference sequence for each evaluated object; (v): Calculate the correlation coefficient, and calculate the correlation coefficient between each comparison sequence and the reference sequence respectively; (vi): Calculate and rank the correlation order. For each evaluation object, calculate the mean value of the correlation coefficient between each indicator and the corresponding element of the reference sequence, so as to reflect the correlation between each evaluation object and the reference sequence.

[0011] In step (iv), the researcher will make specific responses in the following three situations: Scenario 1: When the semantic or syntactic description in the term being studied does not conform to the wishes of the public participants, the researcher removes the semantic or syntactic description from the term. Scenario 2: When the semantic or syntactic description in the term being studied aligns with the wishes of the public participants, the researcher retains that semantic or syntactic description in the term. Scenario 3: When it is uncertain whether the semantic or syntactic description in the term being studied aligns with the wishes of the public participants, the researcher contacts the relevant public participants and provides a guided continuation response to guide them to make further distinctions. Once the public participants' wishes are determined, the researcher then responds to whether the semantic or syntactic description should be retained in the term.

[0012] In step nine, a hierarchical interpretation rule is set for the absolute difference: When the absolute difference between the corresponding elements of the indicator sequence of the evaluation object and the reference sequence is zero, the correlation is zero, indicating that there is no correlation between the two. When the absolute difference between the corresponding elements of the indicator sequence of the evaluation object and the reference sequence is less than or equal to 0.5, it indicates that there is a weak correlation between the two. When the absolute difference between the corresponding elements of the indicator sequence of the evaluation object and the reference sequence is greater than 0.5, it indicates that there is a strong correlation between the two. When the absolute difference between the corresponding elements of the evaluation object's index sequence and the reference sequence is equal to 1, it indicates that there is a perfect correlation between the two.

[0013] This invention improves the objectivity and scientific rigor of quantifying public perception in spatial environment surveys by extracting tense and syntactic features from interview texts and combining them with spatial environmental features from photographs for grayscale correlation analysis. This provides data guidance for matters such as the renovation, reconstruction, and sewage treatment of specific spatial environments. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating steps one through nine of the present invention. Figure 2 This is a flowchart illustrating the analytical method of encoding and decoding semantic elements in interview texts using the theory of combination. Figure 3 It is a schematic diagram of the tree diagram obtained by parsing the syntactic and semantic structure of the text; Figure 4 This is a schematic diagram of a gray-scale correlation analysis model of the coupling degree between spatial environmental elements and semantic elements. Detailed Implementation

[0015] like Figures 1 to 4 As shown, this invention provides a method for analyzing the semantics and element associations of spatial environment based on weighted photography. The subjects involved include researchers of specific spatial environments and the general public. Among the general public, those who participate in spatial environment surveys are public participants.

[0016] Taking a specific characteristic town as an example, the following steps are followed: Step 1: Researchers introduce the basic information of the project to the public, including the research team, research topic, research objectives, research significance, project timeline, and project confidentiality matters, so that the public understands the basic information of the project; Step Two: Researchers recruit public participants who are interested in the space of this distinctive town and are willing to participate in the project; the public registers to participate in the project. Step 3: Researchers record basic information about public participants and provide them with photography training, including clarifying the research topic, how to use photography equipment, and determining the time and scope of photography; researchers and public participants sign confidentiality agreements and authorization letters, and agree on a meeting time and place; Step 4: Each public participant moves freely in a specific street space of the characteristic town, taking photos independently without researcher supervision, and using the photos to show their views and experiential understanding of the quality of the street space; the researchers wait for the public participants to complete their photography. Step 5: After all public participants finish taking photos, both parties will meet at the agreed time and place, and each public participant will meet with the researcher separately; Step Six: During the meeting, public participants select photos they wish to share from the photos taken and explain their photos to the researchers, stating why they took the photos. The explanations should revolve around the research theme. Researchers should listen attentively and refrain from interrupting, commenting on, or guiding the conversation until the public participants have finished speaking. Researchers should only respond to the interview when necessary, such as when the public participants stop speaking, and then provide a guiding response to any unmentioned elements in the photos. Step 7: After completing Step 6 with one public participant, the researcher will proceed to Step 6 with the next public participant, until the researcher has completed Step 6 with all public participants, thus concluding all interviews. Step 8: Researchers compile the information forms, empowered photographs, and interview audio recordings of all public participants, and classify and number the obtained data. Step Nine: Researchers transcribed the interview audio into text, used collocation theory to encode and decode the semantic elements of the interview text, parsed the syntactic and semantic structure of the interview text into a tree diagram, extracted the tense and syntactic features in the language pattern, and constructed a semantic comparison sequence. At the same time, they extracted the spatial environmental element features in the photos through digital image matching technology and constructed a spatial reference sequence. Then, they used a gray-scale correlation analysis model to determine the semantic relationship between the semantics of the spatial environmental elements and the tense, syntax, and semantics in the language pattern, and obtained qualitative graphic data of weighted photography and photo interviews, which represent the public's overall perception of the street space of the characteristic town.

[0017] Through the above-mentioned technical solution, this invention effectively improves the objectivity and scientific nature of public participation in spatial environment perception, and provides data guidance for matters such as the transformation, reconstruction, and sewage treatment of specific spatial environments.

Claims

1. A method for analyzing the semantics and element associations of spatial environment based on empowered photography, involving researchers of specific spatial environments and the general public, with those participating in spatial environment surveys being public participants; characterized by... Follow these steps: Step 1: The public learns about the basic information of the project; the researchers introduce the basic information of the project to the public, including the research team, research topic, research objectives, research significance, project timeline, and project confidentiality matters, etc. Step Two: Public Registration for the Project; Researchers recruit members of the public who are interested in the specific spatial environment and willing to participate in the project; Step 3: Researchers and public participants define their respective rights and obligations; researchers record the basic information of public participants and provide them with photography training, including but not limited to clarifying the research topic, how to use photographic equipment, and determining the time and scope of photography; researchers and public participants sign confidentiality agreements and authorization letters, and agree on a meeting time and place; Step 4: Each public participant takes photographs in a specific spatial environment; Each public participant moves freely in a specific spatial environment and takes photographs independently without being monitored by researchers, using the photos to demonstrate their views on the research topic, as well as their experience and understanding of the research topic related to the specific spatial environment. Step 5: End photography and begin the meeting; All public participants will end their photography session and both parties will meet at the agreed time and place. Each public participant will meet with the researcher separately. Step Six: Selecting and Explaining Photos; During the meeting, public participants select photos they wish to share from the photos taken and decide the order in which they share their experiences and understanding of the research topic; Researchers observe and record the public participants' selection process and the photos they choose, and prepare equipment to record the interviews. Public participants explained their photos to researchers and explained why they took the photos, with the explanations revolving around the research theme. Researchers listened attentively and did not interrupt, comment on, or guide the conversation until the public participants had finished speaking. They only responded to the questions when necessary. Step 7: End the interview; After completing Step 6 with one public participant, the researcher will proceed to Step 6 with the next public participant, until the researcher has completed Step 6 with all public participants, thus ending all interviews. Step 8: Organize the data; Researchers organize the information forms, empowered photographs, and interview audio recordings of all public participants, and classify and number the obtained data. Step Nine: Researchers transcribed the interview audio into text, used collocation theory to encode and decode the semantic elements of the interview text, parsed the syntactic and semantic structure of the interview text into a tree diagram, extracted the tense and syntactic features in the language pattern, and constructed a semantic comparison sequence. At the same time, they extracted the spatial environment element features in the photos through digital image matching technology and constructed a spatial reference sequence. Then, they used a gray-scale correlation analysis model to determine the semantic relationship between the semantics of the spatial environment elements and the tense, syntax, and semantics in the language pattern, and obtained qualitative graphic data of weighted photography and photo interviews, which represent the overall cognition of the public regarding a specific spatial environment.

2. The spatial environment semantics and element association analysis method based on weighted photography according to claim 1, characterized in that: In step six, the interview response under the necessary circumstances includes one of the following four scenarios: Scenario 1: When public participants show a lack of confidence or little interest, researchers can make an attentional response to show their concern and appreciation for the public participants, so as to encourage them to express their true photographic situation and their experience and understanding of the specific spatial environment. Scenario 2: When researchers are unclear about the meaning or attitude of public participants' statements, they can make a clarifying response to confirm the content of the public participants' conversations. Scenario 3: When researchers want to know more details about what public participants are saying, they can make a follow-up questioning response, asking more questions about the content of the public participants' conversations. Scenario 4: When public participants stop talking, researchers can provide a guided continuation response to encourage them to talk more, or researchers can carefully examine the content of the photographs and provide guided continuation responses to the objects in the images that the public participants did not talk about.

3. The spatial environment semantics and element association analysis method based on weighted photography according to claim 1, characterized in that: In step nine, the encoding and decoding of semantic elements in the interview text using the application of grouping theory includes the following steps: (i): Parse the syntactic and semantic structure of the interview text into a tree diagram, starting from the top of the tree diagram of the syntactic and semantic structure of the interview text, that is, from the main clause or the main predicate structure; (ii): Find a term contained in the tree diagram of the semantic structure discussed in the photograph; (iii): Perform word, syntactic and semantic analysis using the methods in the next step, i.e., step (iv); (iv): The specific process for applying combination limit encoding and decoding to the entries is as follows: (a) Subtract the semantic or syntactic description from the tree diagram that the researcher started with; (b) Add semantics or syntax to the tree diagram that the researcher is drawing; (c) Repeat steps (ii) and (iii) to continue adding semantic or syntactic descriptions to the tree diagram until the vocabulary in the tree diagram that the researcher started with is used up; (d) Test whether the tree diagram completed by the researchers is qualified.

4. The spatial environment semantics and element association analysis method based on weighted photography according to claim 1, characterized in that: In step nine, the adoption of the gray-scale correlation analysis model includes the following steps: (i): The analysis sequence is determined according to the evaluation purpose of the space environment elements, including the reference sequence and the comparison sequence; wherein, the reference sequence refers to the real physical information of the space environment element characteristics shown in the photograph, which is determined by digital image matching technology; the comparison sequence is the features that affect the changes of the space environment elements, which is determined by semantic application combination. (II): Determine the indicator data and reference data columns; (iii): Dimensionless processing of indicator data; (iv): Calculate the absolute difference between the corresponding elements of the indicator sequence and the reference sequence for each evaluated object; (v): Calculate the correlation coefficient, and calculate the correlation coefficient between each comparison sequence and the reference sequence respectively; (vi): Calculate and rank the correlation order. For each evaluation object, calculate the mean value of the correlation coefficient between each indicator and the corresponding element of the reference sequence, so as to reflect the correlation between each evaluation object and the reference sequence.

5. The spatial environment semantics and element association analysis method based on weighted photography according to claim 3, characterized in that: In step (iv), the researcher will make specific responses in the following three situations: Scenario 1: When the semantic or syntactic description in the term being studied does not conform to the wishes of the public participants, the researcher removes the semantic or syntactic description from the term. Scenario 2: When the semantic or syntactic description in the term being studied aligns with the wishes of the public participants, the researcher retains that semantic or syntactic description in the term. Scenario 3: When it is uncertain whether the semantic or syntactic description in the term being studied aligns with the wishes of the public participants, the researcher contacts the relevant public participants and provides a guided continuation response to guide them to make further distinctions. Once the public participants' wishes are determined, the researcher then responds to whether the semantic or syntactic description should be retained in the term.

6. The spatial environment semantics and element association analysis method based on weighted photography according to claim 4, characterized in that: In step nine, a hierarchical interpretation rule is set for the absolute difference: When the absolute difference between the corresponding elements of the indicator sequence of the evaluation object and the reference sequence is zero, the correlation is zero, indicating that there is no correlation between the two. When the absolute difference between the corresponding elements of the indicator sequence of the evaluation object and the reference sequence is less than or equal to 0.5, it indicates that there is a weak correlation between the two. When the absolute difference between the corresponding elements of the indicator sequence of the evaluation object and the reference sequence is greater than 0.5, it indicates that there is a strong correlation between the two. When the absolute difference between the corresponding elements of the evaluation object's index sequence and the reference sequence is equal to 1, it indicates that there is a perfect correlation between the two.

Images