Food sensory evaluation device and method capable of synchronously collecting vision and mouse operation

Abstract

The invention discloses a food sensory evaluation device and method capable of synchronously collecting vision and mouse operation, and belongs to the technical field of food sensory digital evaluation. The device comprises a telescopic experiment table, a display screen, a visual behavior capture instrument, a food heat preservation pad, a mouse, a synchronous control unit and the like, and can synchronously collect eye movement tracks and mouse operation data of a testee under the condition of constant temperature. The method comprises the steps of constant-temperature control, eye movement collection, mouse operation, synchronous collection, data association verification and result output, and the data validity is verified by calculating the matching degree of pupil staying time and mouse operation duration. According to the method, temperature fluctuation and subjective deviation are effectively overcome, objectivity, accuracy and repeatability of food sensory evaluation are improved, and the method is suitable for multiple application scenes such as food research and development, quality control and market research.

Description

Technical Field

[0001] This invention belongs to the field of digital sensory evaluation technology for food, specifically relating to a device and method for simultaneously acquiring visual and mouse operation data for food sensory evaluation. Background Technology

[0002] Sensory evaluation of food is a crucial step in product development, quality control, and market research, directly influencing formula adjustments, shelf-life settings, and consumer positioning. Currently, mainstream sensory evaluation relies primarily on manual scoring sheets or electronic questionnaires. However, this process is susceptible to interference from factors such as ambient temperature fluctuations, sample temperature drops, and subjective preferences of participants, leading to poor data repeatability and significant bias. Especially in multi-round, multi-sample comparisons, temperature changes can significantly alter flavor and texture, rendering evaluation results incomparable.

[0003] To improve objectivity, related technologies have attempted to introduce eye-tracking into sensory laboratories, indirectly reflecting subjects' true preferences for samples by recording visual behavioral parameters such as pupil dwell time and gaze trajectory. However, most existing methods lack constant temperature control of the samples themselves, and there is a lack of strict time synchronization between eye-tracking data and sensory stimuli. At the same time, some methods still require input via mouse or keyboard, and this additional operation further interferes with the subjects' natural responses, affecting the accuracy of the data.

[0004] Therefore, there is an urgent need for a sensory evaluation method with a simple structure that can simultaneously capture pupil movement trajectories under constant sample temperature conditions, in order to reduce subjective bias and improve the repeatability and reliability of evaluation results. Summary of the Invention

[0005] [Technical Issues] Existing methods for sensory evaluation of food lack constant temperature control and synchronization with eye-tracking data, resulting in sample temperature drift, large data time deviation, high subjective bias, and insufficient repeatability and reliability of results.

[0006] [Technical Solution] To address the above issues, this invention provides a food sensory evaluation device and method that simultaneously acquires visual and mouse operation data. While maintaining a constant sample temperature, it can achieve dual verification based solely on eye movement trajectory and sensory scores, significantly improving the objectivity, accuracy, and reliability of sensory evaluation.

[0007] In a first aspect, the present invention provides a food sensory evaluation device for simultaneously acquiring visual and mouse operation data, comprising a retractable experimental table, the retractable experimental table adopting a bidirectional synchronous sliding rail structure, the retraction direction being the length direction, the retraction range being 600nm-1100nm, and locking after retraction via a knob screw; a display screen is fixed to the tabletop of the retractable experimental table, and a visual behavior capture device is fixed below the display screen; the visual behavior capture device and screen calibration device are integrated into a single design, and are mounted together with a high-definition infrared eye-tracking camera below the display screen, the optical axis of the high-definition infrared eye-tracking camera being perpendicular to the upper surface of the food insulation pad with an embedded PID closed-loop temperature control module, and intersecting with the geometric center of the upper surface; the PID closed-loop... The temperature control module consists of a temperature sensor, a solid-state relay, a PID controller, and a thin-film heating element. The overall thickness of the food insulation pad is ≤8mm. After being embedded in the tabletop of the extendable experimental table, it remains flush with the tabletop. The bearing surface size is adapted to conventional sample containers. It is fixed by a slot positioning and is located directly in front of the display screen. The left side of the extendable experimental table has a faucet and a sink, and the right side has a mouse. A 180-degree rotatable posture corrector is located in the center of the tabletop. The 180-degree rotatable posture corrector has a horizontal axis for rotation and is magnetically fixed after rotation. The corrector body is Y-shaped. Foldable fan-shaped partitions are hinged to the left and right sides of the tabletop of the extendable experimental table. The device also includes a synchronization control unit with a built-in hardware clock chip. The synchronization control unit consists of an ARM main control board, a hardware clock chip, and a power management chip, all integrated inside the visual behavior capture device. The synchronization control unit is wirelessly connected to both the visual behavior capture device and the mouse.

[0008] Optionally, the sampling frequency of the high-definition infrared eye-tracking camera is not less than 120 Hz, and the screen calibration device has nine calibration points with a calibration error of no more than 0.5 degrees of viewing angle.

[0009] Optionally, the resolution of the display screen is not less than 1920×1080, and the positioning error of the mouse click position is not greater than 1mm.

[0010] Secondly, the present invention provides a method for sensory evaluation of food, comprising: Temperature control steps: The food insulation pad is used to perform PID closed-loop temperature control on the sample to be tested, so that the sample temperature is maintained at the set value, and temperature data is collected. Eye-tracking acquisition steps: The electronic sensory evaluation scale is displayed on the screen, and the visual behavior capture device captures the eye-tracking trajectory data of the subject's pupils on the interest area of ​​the electronic sensory evaluation scale in real time; the eye-tracking trajectory data includes pupil dwell time data, pupil fixation frequency data, and pupil gaze trajectory data; Mouse operation steps: Participants click and score the area of ​​interest on the electronic sensory evaluation scale using the mouse, and the mouse operation data is recorded; the mouse operation data includes mouse click location data and mouse operation duration data; Synchronous acquisition steps: A unified clock signal is generated by the hardware clock chip in the synchronous control unit. This signal is simultaneously sent to the food warming pad, the visual behavior capture device, and the mouse to trigger the acquisition of temperature data, the acquisition of eye movement trajectory data, and the recording of mouse operation data, thereby ensuring that the time error of these three types of data does not exceed 10 milliseconds. Data association verification steps: Calculate the pupil dwell time in the eye movement trajectory data within the same region of interest. The mouse operation duration in the mouse operation data Matching degree And based on the matching degree Generate verification conclusions; wherein, the matching degree The calculation formula is:

[0011] in, Represents pupil dwell time; This represents the duration of mouse operation; Output steps: Output a complete dataset containing temperature data, eye movement trajectory data, mouse operation data, and verification conclusions.

[0012] Optionally, the electronic sensory evaluation scale is obtained through the following methods: The first step is to collect historical sensory evaluation feature words of the samples to be tested and digitize them to establish a preliminary feature word library; The second step involves collecting eye movement data of the subjects in response to electronic feature words using the visual behavior capture device, and statistically analyzing the distribution of pupil fixation duration and fixation frequency in each feature word region. The third step is to select the feature words that the subjects pay more attention to based on the above distribution of fixation duration and fixation frequency to improve the preliminary feature word library and obtain the target feature word set. The fourth step is to select feature words from the target feature word set as sensory evaluation indicators and generate an electronic sensory evaluation scale with the sensory evaluation indicators and corresponding scoring options.

[0013] Optionally, in the constant temperature control step, a PID closed-loop temperature control method is used to control the temperature of the sample to be tested within the range of 20℃ to 60℃, the error between the controlled temperature and the set value does not exceed 1℃, and the sample reaches thermal equilibrium within 30s.

[0014] Optionally, the area of ​​interest refers to the 1 cm × 1 cm operable area where the scoring option corresponding to each sensory evaluation index is located in the electronic sensory evaluation scale.

[0015] Optionally, in the data association verification step, the matching degree is used as a basis. The method for generating verification conclusions is as follows: when the calculated matching degree... When the matching degree is high, confirm that the current result is valid; when the matching degree is high, confirm that the current result is valid. When, recalculate the matching degree until Or it may be manually marked as abnormal.

[0016] Optionally, the complete dataset output in the results output step is saved as a CSV or Excel file; the file can be imported into SPSS, R, or other mainstream statistical analysis software for analysis and statistics; the file contains fields such as: sample number, temperature curve, etc. , , Values ​​and verification conclusions.

[0017] Thirdly, the present invention provides a food sensory evaluation system, comprising: The temperature control module is used to control the temperature of the sample under test and collect temperature data. The eye-tracking acquisition module is used to collect eye movement trajectory data from the subjects. The mouse operation module is used to collect mouse operation data from the subjects. The synchronization trigger module is used to generate a synchronization clock signal to ensure the time synchronization between the eye movement data and the mouse operation data; The data association verification module is used to receive the above data and calculate the matching degree between pupil dwell time and mouse operation duration within the same region of interest. The results output module is used to output the complete dataset.

[0018] [Beneficial Effects] 1. In the constant temperature control step, the present invention ensures that the temperature of the sample remains constant in multiple rounds of evaluation through the food insulation pad and its embedded PID closed-loop temperature control module, effectively eliminating sensory differences caused by temperature drift, and significantly improving the comparability and repeatability of evaluation results between different batches and different samples.

[0019] 2. In the eye-tracking acquisition step, mouse operation step, and synchronous triggering step, this invention introduces a high-definition infrared eye-tracking camera arranged vertically and coaxially and a synchronous acquisition mechanism for mouse click operation. It records the pupil's gaze behavior in the region of interest at a sampling rate of not less than 120 Hz and achieves millisecond-level synchronization with the mouse click behavior through a unified clock signal, thereby realizing dual data capture and correlation verification of "implicit visual behavior" and "explicit operational behavior".

[0020] 3. In the data association verification step, this invention proposes a matching degree calculation and secondary confirmation mechanism, which uses formulas to determine data consistency and effectively identify and correct subjective deviations caused by distraction, misoperation, etc.

[0021] 4. The data output in the result output step of this invention is complete and highly compatible. The result set includes isothermal curves, eye-tracking trajectories, mouse operation data, and verification conclusions. It supports export in both CSV and Excel formats and can be directly connected to statistical analysis tools such as SPSS and R, thus improving data processing efficiency.

[0022] 5. The device of this invention integrates constant temperature control, visual capture, operation input, environmental isolation, and cleaning functions. Its compact design makes the operation process highly consistent with traditional sensory evaluation, eliminating the need for complex training of evaluators. It can be widely used in food company R&D laboratories, university sensory science institutes, and third-party market research agencies, perfectly covering various application scenarios such as new product flavor adjustment, packaging design and optimization, and consumer preference testing. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a structural diagram of the device provided by the present invention.

[0025] Figure 2 A front view of the device provided by the present invention.

[0026] Figure 3 A side view of the device provided by the present invention.

[0027] Figure 4 A top-desk view of the device provided by the present invention.

[0028] Figure 5 A top view of the folded partition of the device provided by the present invention in the open state.

[0029] Figure 6 The diagram shows the display screen and visual tracker of the device provided by the present invention.

[0030] Figure 7 A diagram of a faucet and a sink for the device provided by the present invention.

[0031] Figure 8 A diagram of a 180-degree flip-up posture corrector for the device provided by the present invention.

[0032] Figure 9 The original feature word map for eye movement observation provided by this invention.

[0033] Figure 10 The eye-tracking observation heatmap provided for this invention.

[0034] Figure 11 The eye-tracking visual observation heatmap provided by this invention.

[0035] Figure 12 A mouse click heatmap provided for this invention.

[0036] Figure 13 Radar chart of milk and coffee packaging preferences provided by the present invention.

[0037] Figure 14 This is a comparison image of milk color recognition under DCV and MCV modes provided by the present invention.

[0038] Figure 15 Radar chart of taste and preference for jasmine tea samples provided by the present invention.

[0039] Figure 16 PCA principal component analysis diagram showing the correlation between sensory attributes and consumer preferences of jasmine tea samples provided by this invention.

[0040] Figure label: Figure 1 and Figure 2 In the middle: 1-Extendable experimental table, 2-Display screen, 3-Visual behavior capture device, 4-180-degree flip-up posture corrector, 5-Faucet, 6-Water sink, 7-Mouse, 8-Foldable fan-shaped partition, 9-Food insulation pad. Detailed Implementation

[0041] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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.

[0042] Example 1: like Figures 1 to 8 As shown, this embodiment provides a food sensory evaluation device that simultaneously acquires visual and mouse operation data, including: The telescopic experimental table (1) adopts a two-way synchronous slide rail structure. The telescopic direction is the length direction, and the telescopic range is 600nm-1100nm. After telescopic extension, it is locked by a knob screw. A display screen (2) is installed in the center of the desktop of the extendable experimental table (1). The display screen (2) has a resolution of not less than 1920×1080. A visual behavior capture device (3) is fixed below the display screen (2). A 180-degree flip-up posture corrector (4) is placed on the table directly in front of the subject. The flip axis is a horizontal axis. After flipping into place, it can be placed against the subject's head and fixed in position with magnetic attraction to ensure the accuracy of data collection by the visual behavior capture device (3). The main body of the 180-degree flip-up posture corrector (4) is "Y" shaped. The left side of the experimental table integrates a faucet (5) and a sink (6) for subjects to rinse their mouths between different sample evaluation intervals, reducing the interference of residual oral flavors on subsequent tests; The mouse (7) is placed on the right side of the desktop. The positioning error of the mouse (7) is ≤1mm, which makes it easy for the subject to operate. The overall thickness of the food insulation pad (9) is ≤8mm. It is embedded in the table and flush with the table. The size of the bearing surface is adapted to conventional sample containers. The fixing method is slot positioning and it is located in front of the display screen. It has a PID closed-loop temperature control module embedded inside, which can keep the sample temperature constant, eliminate the influence of temperature on taste, and further improve the reliability of data. The PID closed-loop temperature control module consists of a temperature sensor, a solid-state relay, a PID controller and a thin film heating element. The extendable experimental table (1) has foldable fan-shaped partitions (8) hinged on both sides. When unfolded, they can form physical isolation to avoid interference from the color, shape and smell of adjacent samples to the test subjects.

[0043] The visual behavior capture device (3) and the screen calibration device are integrated and installed together with the high-definition infrared eye-tracking camera below the display screen (2); the optical axis of the high-definition infrared eye-tracking camera is perpendicular to the upper surface of the food insulation pad (9) and intersects with the geometric center of the upper surface, and the high-definition infrared eye-tracking camera has a sampling frequency of not less than 120 Hz, and the screen calibration device has a 9-point calibration function and the calibration error does not exceed 0.5 degrees of viewing angle.

[0044] The device also includes a synchronization control unit, which consists of an ARM main control board, a hardware clock chip and a power management chip, all integrated inside the visual behavior capture device; the synchronization control unit is electrically connected to the visual behavior capture device (3) and the mouse (7) respectively; and the synchronization control unit generates a synchronization clock signal through the hardware clock chip so that the time error between the visual tracking data and the mouse operation data does not exceed 10 milliseconds.

[0045] The display screen (2) in the device is used to present sensory evaluation indicators, the mouse (7) is used to record the subject's click selection and operation duration, and the visual behavior capture device (3) is used to capture in real time the subject's pupils stay in the sensory evaluation indicator area presented on the display screen, the number of gazes, and the gaze trajectory.

[0046] Example 2 This embodiment provides a food sensory evaluation method, which uses a food sensory evaluation device that simultaneously collects visual and mouse operations as provided in Embodiment 1, including: Temperature control steps: The temperature of the sample to be tested is controlled by PID closed-loop control through a food insulation pad, and the temperature of the sample to be tested is controlled within the range of 20℃ to 60℃. The error between the controlled temperature and the set value shall not exceed 1℃, and the sample shall reach thermal equilibrium within 30s and the temperature data shall be collected. Eye-tracking data acquisition steps: The electronic sensory evaluation scale is displayed on the screen, and the visual behavior capture device captures the eye movement trajectory data of the subject's pupils on the interest area of ​​the electronic sensory evaluation scale in real time; the eye movement trajectory data includes pupil dwell time data, pupil fixation number data, and pupil gaze trajectory data. Mouse operation steps: Participants clicked and scored the areas of interest on the electronic sensory evaluation scale using a mouse, and the mouse operation data was recorded; the mouse operation data included mouse click location data and mouse operation duration data; Synchronous acquisition steps: A unified clock signal is generated by the hardware clock chip in the synchronous control unit. This signal is simultaneously sent to the food warming pad, the visual behavior capture device, and the mouse to trigger the acquisition of temperature data, eye movement trajectory data, and mouse operation data, thereby ensuring that the time error of these three types of data does not exceed 10 milliseconds. Data association verification steps: Calculate pupil dwell time in eye movement trajectory data within the same region of interest. Mouse operation duration in mouse operation data Matching degree And based on the matching degree Generate validation conclusions; among which, the matching degree The calculation formula is:

[0047] in, Represents pupil dwell time; This represents the duration of mouse operation; When the calculated matching degree When the matching degree is high, confirm that the current result is valid; when the matching degree is high, confirm that the current result is valid. When, recalculate the matching degree until Or manually marked as abnormal; Results Output Steps: Output a complete dataset containing temperature data, eye-tracking data, mouse operation data, and validation conclusions. Save the dataset as a CSV or Excel file. This file can be imported into SPSS, R, or other mainstream statistical analysis software for analysis. The file content includes the following fields: sample number, temperature curve, etc. , , Values ​​and verification conclusions.

[0048] Furthermore, the electronic sensory evaluation scale was obtained through the following methods: The first step is to collect historical sensory evaluation feature words of the samples to be tested and digitize them to establish a preliminary feature word library; The second step involves collecting eye movement data of the subjects in response to electronic feature words using the visual behavior capture device, and statistically analyzing the distribution of pupil fixation duration and fixation frequency in each feature word region. The third step is to select the feature words that the subjects pay more attention to based on the above distribution of fixation duration and fixation frequency to improve the preliminary feature word library and obtain the target feature word set. The fourth step is to select feature words from the target feature word set as sensory evaluation indicators and generate an electronic sensory evaluation scale with the sensory evaluation indicators and corresponding scoring options.

[0049] Furthermore, the area of ​​interest refers to the 1 cm × 1 cm operable area where the scoring option corresponding to each sensory evaluation indicator is located in the electronic sensory evaluation scale.

[0050] Example 3 This embodiment provides a method for determining subjects' preferences for milk and coffee packaging, aiming to identify the optimal combination of color (red, green, blue, brown, white) and geometric shape (circle, triangle, quadrilateral, pentagon, hexagon) for milk and coffee packaging. The specific implementation steps are as follows: 1. System preparation: Start the computer, run the software system, and input food information such as milk and coffee, subject information such as gender and age, and evaluation indicators such as color and geometric outline through the data acquisition module.

[0051] 2. Visual Feature Word Identification: Historically collected packaging evaluation features, such as "eye-catching" and "harmonious," were compiled into an electronic version. A feature word library was established using eye-tracking observation of the original feature word map. Eye-tracking heatmaps were then used to further visualize the distribution of participants' attention to each feature word. (See the original feature word map.) Figure 9 Eye movement observation thermal map Figure 10 .

[0052] 3. Sensory feature word determination: Select relevant feature words from the word library, create an electronic feature word list and an electronic questionnaire scale, with scores set from 1 to 5 points, and each score option marked as a 1cm*1cm area of ​​interest.

[0053] 4. Subject Preparation: Subjects sat in front of the monitor, used a posture corrector to fix their heads, and completed a 9-point calibration. The experimenter informed the subjects that they could rinse their mouths with water from the tap and sink between samples to avoid interference from residual flavors.

[0054] 5. Data Acquisition: (1) Feature word attention data collection: An electronic feature word list is displayed, and the subject views it for 3-5 minutes. An eye-tracking device records the pupil fixation time, number of fixations, and number of re-applications at each feature word, such as... Figure 9 and Figure 10 The data is then transmitted to the data acquisition module for storage.

[0055] (2) Scoring data collection: The electronic questionnaire is displayed. The subjects score according to their preferences by clicking the score with the mouse. The data collection module records the mouse click position and selection time simultaneously. The eye tracker records the eye movement data of each area of ​​interest. The two sets of data are bound and stored.

[0056] 6. Data Processing: The data processing module generates an eye-tracking visual observation heatmap (see...). Figure 11 ) and mouse click heatmap (see Figure 12 The consistency between the two is compared. If the matching degree M between eye movement dwell time and mouse operation time in the same area of ​​interest is ≤0.3 (e.g., M=0.05 for the "red-circle" combination), the result is considered valid; if M>0.3 (e.g., M=0.42 for the "red-triangle" combination), a secondary confirmation process is automatically triggered until the recalculated matching degree M′≤0.3 or is marked as abnormal data by the experimenter.

[0057] 7. Results Presentation: The overall results are presented as a radar chart of preferences for milk and coffee packaging, see below. Figure 13 The results are presented in red, representing eye-tracking results, and in blue, representing questionnaire results.

[0058] The conclusion of this embodiment is that white-circle, red-circle, and brown-circle are the best preferred combinations for both milk and coffee packaging; the preference for geometric outlines is significantly lower in the coffee system than in the milk system, indicating that there are category differences in visual preferences.

[0059] Example 4 This embodiment focuses on constructing a sensory detection system for milk color quality, specifically verifying the effectiveness of multicolor feature words (MCV) compared to traditional two-color feature words (DCV) in identifying milk with substandard color. The specific steps are as follows: 1. System and subject preparation: Same as in Example 3.

[0060] 2. Evaluation index settings: DCV mode uses two types of feature words, white and yellow; MCV mode uses multiple feature words such as LY (light yellow), Y (yellow), LGy (light greenish yellow).

[0061] 3. Data Collection: A multi-colored milk sample containing substandard products was presented. Subjects completed eye-tracking observation and mouse scoring operations in DCV and MCV modes, and the relevant data were recorded simultaneously.

[0062] 4. Data processing: Perform a significant difference analysis on the eye-tracking data and questionnaire data obtained under different modes.

[0063] The results are as follows Figure 14 The diagram shows that a, ab, and b are markers of significant differences, with different letters indicating significant differences. Figures A and B compare different color feature words under short fixation, long fixation, and questionnaire scores (short fixation and long fixation are eye movement indicators, representing the degree of attention to a certain area, not the total experimental time). In other words, under DCV mode, traditional questionnaires can hardly identify color differences, while eye movement analysis can identify 20%-30% of non-compliant samples. Under MCV mode, the questionnaire recognition rate reaches 42%, and the eye movement analysis recognition rate is further improved to 70%-80%, indicating that MCV mode has higher sensitivity and effectiveness in identifying non-compliant products due to color.

[0064] Example 5 This embodiment demonstrates, through a comparative sensory evaluation experiment of jasmine tea, the advantages of the "constant temperature + eye movement" mode of the present invention in improving the objectivity of the evaluation compared to traditional room temperature questionnaires. The specific implementation is as follows: 1. System and subject preparation: Same as in Example 3, with the following additional settings: Activate the food temperature control pad and keep the temperature constant at 25℃ (the accuracy error of the PID closed-loop temperature control module does not exceed 1℃); prepare three jasmine tea samples, including Sample 1 (low concentration sugar-free), Sample 2 (high concentration sugar-free), and Sample 3 (high sugar).

[0065] 2. Evaluation index setting: Evaluation indexes such as sweet aroma, fruity aroma, tea aroma, jasmine aroma, sophistication, and likability were set for all samples, with a scoring level of 1-9 points.

[0066] 3. Experimental procedure: The sample is placed on a constant temperature pad and reaches thermal equilibrium within 30 seconds; the screen calibration device performs nine-point calibration before data acquisition, with a calibration error of no more than 0.5 degrees of visual angle; the subject observes the sample and clicks to score, the system simultaneously collects eye movement data and operation data, and calculates the matching degree M in real time. If M > 0.3, a secondary confirmation is initiated.

[0067] Combination Figure 15 (The left side shows the questionnaire results, and the right side shows the eye-tracking results of this embodiment.) Figure 16 (From top to bottom are the questionnaire results and the eye-tracking results of this embodiment) As shown: (1) Temperature stability: The temperature drift under the constant temperature setting of the present invention does not exceed 0.5ºC, while the temperature drift under normal temperature is about 3.2ºC; (2) Data consistency: Under the "constant temperature + eye-tracking" mode of the present invention, the correlation between sweetness and likability reached r=0.78, which is significantly higher than the r=0.34 of the traditional normal temperature questionnaire; (3) Subjective bias control: Under the constant temperature setting of the present invention, the abnormal scoring trigger rate is only 5%, while the abnormal scoring trigger rate under normal temperature setting is 28%; (4) Feature discrimination: The "constant temperature + eye-tracking" mode of the present invention increases the intensity of revealing the negative correlation between "sweetness" and "high-end feeling" by 129%.

[0068] The "constant temperature + eye movement" synchronous food sensory evaluation method of the present invention can effectively suppress temperature fluctuations and subjective arbitrariness, significantly improve the objectivity of jasmine tea sensory evaluation, and provide reliable decision support for the optimization of low-sugar formulas.

[0069] In summary, by implementing the above embodiments, the present invention demonstrates significant advantages in reducing temperature drift, controlling subjective bias, improving data repeatability and scenario adaptability, and can be widely applied in multiple fields such as food research and development, market research and packaging optimization.

[0070] Example 6 This embodiment provides a food sensory evaluation system for executing the food sensory evaluation method provided in Embodiment 2, including: The temperature control module is used to control the temperature of the sample under test and collect temperature data. The eye-tracking acquisition module is used to collect eye movement trajectory data from the subjects. The mouse operation module is used to collect mouse operation data from the subjects. The synchronization trigger module is used to generate a synchronization clock signal to ensure the time synchronization between the eye movement data and the mouse operation data; The data association verification module is used to receive the above data and calculate the matching degree between pupil dwell time and mouse operation duration within the same region of interest. The results output module is used to output the complete dataset.

[0071] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A food sensory evaluation device that simultaneously acquires visual and mouse operation data, characterized in that, The system includes a retractable experimental table with a bidirectional synchronous sliding rail structure. The retraction direction is along the length, and the retraction range is 600nm-1100nm. After extension, it is locked using a knob and screw. A display screen is fixed to the tabletop, and a visual behavior capture device is fixed below the display screen. The visual behavior capture device is integrated with a screen calibration device and mounted below the display screen along with a high-definition infrared eye-tracking camera. The optical axis of the high-definition infrared eye-tracking camera is perpendicular to the upper surface of the food insulation pad with an embedded PID closed-loop temperature control module, and intersects with the geometric center of this upper surface. The PID closed-loop temperature control module consists of a temperature sensor and a solid-state relay. The system consists of a PID controller and a thin-film heating element; the overall thickness of the food insulation pad is ≤8mm, and it remains flush with the tabletop after being embedded in the retractable experimental table. The bearing surface size is adapted to conventional sample containers, and it is fixed by a slot positioning and located directly in front of the display screen; the left side of the retractable experimental table is equipped with a faucet and a sink, and the right side is equipped with a mouse; a 180-degree rotatable posture corrector is located in the center of the tabletop; the 180-degree rotatable posture corrector has a horizontal axis for rotation and is magnetically fixed after being rotated into place. The corrector body is "Y" shaped; foldable fan-shaped partitions are hinged to the left and right sides of the tabletop of the retractable experimental table; The device also includes a synchronization control unit with a built-in hardware clock chip. The synchronization control unit consists of an ARM main control board, a hardware clock chip, and a power management chip, all integrated inside the visual behavior capture device. The synchronization control unit is wirelessly connected to both the visual behavior capture device and the mouse.

2. The apparatus according to claim 1, characterized in that, The sampling frequency of the high-definition infrared eye-tracking camera is no less than 120 Hz, and the screen calibration device has nine calibration points with a calibration error of no more than 0.5 degrees of viewing angle.

3. The apparatus according to claim 1, characterized in that, The resolution of the display screen is not less than 1920×1080, and the positioning error of the mouse click position is not greater than 1mm.

4. A method for sensory evaluation of food, characterized in that, The food sensory evaluation method, applied to a food sensory evaluation device for simultaneously acquiring visual and mouse operations as described in any one of claims 1 to 3, comprises: Temperature control steps: The food insulation pad is used to perform PID closed-loop temperature control on the sample to be tested, so that the sample temperature is maintained at the set value, and temperature data is collected. Eye-tracking acquisition steps: The electronic sensory evaluation scale is displayed on the screen, and the visual behavior capture device captures the eye-tracking trajectory data of the subject's pupils on the interest area of ​​the electronic sensory evaluation scale in real time; the eye-tracking trajectory data includes pupil dwell time data, pupil fixation frequency data, and pupil gaze trajectory data; Mouse operation steps: Participants click and score the area of ​​interest on the electronic sensory evaluation scale using the mouse, and the mouse operation data is recorded; the mouse operation data includes mouse click location data and mouse operation duration data; Synchronous acquisition steps: A unified clock signal is generated by the hardware clock chip in the synchronous control unit. This signal is simultaneously sent to the food warming pad, the visual behavior capture device, and the mouse to trigger the acquisition of temperature data, the acquisition of eye movement trajectory data, and the recording of mouse operation data, thereby ensuring that the time error of these three types of data does not exceed 10 milliseconds. Data association verification steps: Calculate the pupil dwell time in the eye movement trajectory data within the same region of interest. The mouse operation duration in the mouse operation data Matching degree And based on the matching degree Generate verification conclusions; wherein, the matching degree The calculation formula is: in, Represents pupil dwell time; This represents the duration of mouse operation; Output steps: Output a complete dataset containing temperature data, eye movement trajectory data, mouse operation data, and verification conclusions.

5. The method according to claim 4, characterized in that, The electronic sensory evaluation scale was obtained through the following methods: The first step is to collect historical sensory evaluation feature words of the samples to be tested and digitize them to establish a preliminary feature word library; The second step involves collecting eye movement data of the subjects in response to electronic feature words using the visual behavior capture device, and statistically analyzing the distribution of pupil fixation duration and fixation frequency in each feature word region. The third step is to select the feature words that the subjects pay more attention to based on the above distribution of fixation duration and fixation frequency to improve the preliminary feature word library and obtain the target feature word set. The fourth step is to select feature words from the target feature word set as sensory evaluation indicators and generate an electronic sensory evaluation scale with the sensory evaluation indicators and corresponding scoring options.

6. The method according to claim 4, characterized in that, In the constant temperature control step, the temperature of the sample to be tested is controlled within the range of 20℃ to 60℃ using a PID closed-loop temperature control method. The error between the controlled temperature and the set value does not exceed 1℃, and the sample reaches thermal equilibrium within 30s.

7. The method according to claim 4, characterized in that, The area of ​​interest refers to the 1 cm × 1 cm operable area where the scoring option corresponding to each sensory evaluation index is located in the electronic sensory evaluation scale.

8. The method according to claim 4, characterized in that, In the data association verification step, based on the matching degree The method for generating verification conclusions is as follows: when the calculated matching degree... When the matching degree is high, confirm that the current result is valid; when the matching degree is high, confirm that the current result is valid. When, recalculate the matching degree until Or it may be manually marked as abnormal.

9. The method according to claim 4, characterized in that, The complete dataset output in the results output step is saved as a CSV or Excel file; this file can be imported into SPSS, R, or other mainstream statistical analysis software for analysis and statistics; the file contains fields such as: sample number, temperature curve, etc. , , Values ​​and verification conclusions.

10. A food sensory evaluation system, characterized in that, The system for performing the food sensory evaluation method as described in any one of claims 4 to 9 includes: The temperature control module is used to control the temperature of the sample under test and collect temperature data. The eye-tracking acquisition module is used to collect eye movement trajectory data from the subjects. The mouse operation module is used to collect mouse operation data from the subjects. The synchronization trigger module is used to generate a synchronization clock signal to ensure the time synchronization between the eye movement data and the mouse operation data; The data association verification module is used to receive the above data and calculate the matching degree between pupil dwell time and mouse operation duration within the same region of interest. The results output module is used to output the complete dataset.

Images