Visual-oriented multi-source perception evaluation system for concrete truck transportation quality

Abstract

This invention relates to the field of concrete transportation, addressing the challenges of perceiving subtle changes in internal temperature, segregation, and surface bleeding during concrete transport. Specifically, it focuses on a vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality, comprising a visual acquisition module, an infrared temperature monitoring module, a material dynamic analysis module, a visual dynamic analysis module, and a comprehensive early warning module. This invention divides the concrete inside the mixer into zones and collects temperature parameters, then calculates the rate of temperature change, determines uneven hydration heat, and predicts the initial setting time window. Furthermore, it extracts the degree of aggregate exposure and the area of ​​surface bleeding through visual equipment, enabling quantitative grading of quality assessment. Early warnings are generated based on various triggering conditions such as temperature anomalies, initial setting risk, aggregate exposure, and surface bleeding, achieving multi-level and multi-type refined early warning capabilities.

Description

Technical Field

[0001] This invention relates to the field of concrete transportation, specifically to a vision-based multi-source perception and assessment system for the transportation quality of concrete mixer trucks. Background Technology

[0002] During transportation, the properties of concrete are in a state of dynamic change. Affected by multiple factors such as traffic conditions, waiting time, ambient temperature and humidity, and the rotation of the tank, problems such as excessively rapid slump loss, segregation, bleeding, and even abnormal initial setting are very likely to occur. Traditional quality management models rely on two-point control, namely factory inspection and on-site sampling. They lack continuous and objective data monitoring for this critical dynamic process of transportation. The judgment of the state of concrete in transit is highly dependent on the personal experience of the driver, and rough estimates are made by tapping the tank and listening to the sound. This is neither scientific nor accurate, and the quality assessment has serious lag and randomness. Once the concrete arrives at the site and fails to meet the workability requirements, it often faces construction difficulties, secondary adjustments, returns, or even abandonment. This not only causes huge waste of materials and energy, but may also lead to project delays, soaring costs, and quality disputes.

[0003] Currently, although attempts have begun to introduce intelligent monitoring systems into the concrete transportation process, the following problems still exist: First, it is difficult to detect the risk of initial setting caused by changes in the internal temperature of concrete; Secondly, it is impossible to determine apparent quality issues such as aggregate exposure or surface bleeding. In addition, conventional systems mostly provide alarms after the fact, lacking a process from perception to assessment to early warning, making it difficult to achieve dynamic tracking and tiered early warning of transportation quality.

[0004] Therefore, how to perform collaborative analysis of visual features and temperature features within the same data flow framework, and generate graded early warning signals based on multi-dimensional evaluation results, is a problem that urgently needs to be solved in this field. Summary of the Invention

[0005] This invention involves dividing the concrete inside the tank into zones and collecting temperature parameters to calculate the rate of temperature change, determine uneven hydration heat, and predict the initial setting time window. In addition, it uses visual equipment to extract the degree of aggregate exposure and the area of ​​surface bleeding, enabling quantitative grading of quality assessment. It generates early warnings based on various triggering conditions such as temperature anomalies, initial setting risk, aggregate exposure, and surface bleeding, achieving multi-level and multi-type refined early warning capabilities. This invention proposes a vision-driven multi-source perception and assessment system for the quality of concrete mixer truck transportation.

[0006] The objective of this invention can be achieved through the following technical solution: a vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality, comprising a vision acquisition module, an infrared temperature monitoring module, a material dynamic analysis module, a vision dynamic analysis module, and a comprehensive early warning module; The visual acquisition module can acquire images of concrete inside the tanker truck through a visual acquisition device, and perform preliminary processing on the acquired concrete images to generate a main data stream. The infrared temperature monitoring module acquires infrared images of the concrete inside the tanker truck using infrared devices, converts the infrared acquisition results into temperature data, divides the concrete inside the tanker truck into regions, generates concrete temperature zones based on the region division results, and imports the temperature zones and corresponding temperature data into the main data stream. The material dynamic analysis module acquires the main data stream, performs temperature analysis based on the main data stream, generates data on concrete temperature distribution characteristics and temperature change trends, and imports the analysis results into the main data stream. The visual dynamic analysis module acquires the main data stream, performs dynamic analysis based on the concrete images, obtains the concrete condition assessment results, and imports the concrete assessment results into the main data stream. The integrated early warning module acquires concrete condition assessment results and temperature analysis results. When any result exceeds the corresponding preset threshold, it generates a multi-level early warning signal and pushes the early warning signal and related data.

[0007] In a preferred embodiment of the present invention, the method for preprocessing the concrete image after acquiring it by the visual acquisition module includes: The original image is preprocessed by grayscale conversion, median filtering for noise reduction, and histogram equalization. The preprocessed images are analyzed to obtain image features including the degree of concrete aggregate exposure and surface water bleeding areas. Add timestamps to image features to form an image feature sequence as the main data stream.

[0008] In a preferred embodiment of the present invention, the infrared temperature monitoring module converts the acquired thermal radiation image into actual temperature values ​​through a calibration curve, and generates a temperature matrix; The concrete area is divided according to the mixing drum of the tanker truck, resulting in multiple temperature zones; The infrared temperature monitoring module calculates the average temperature, maximum temperature, and temperature distribution uniformity of each temperature zone to obtain temperature data, and adds a timestamp to the temperature data to generate time-series temperature data. Each temperature zone and the corresponding time-series temperature data are imported into the main data stream.

[0009] In a preferred embodiment of the present invention, the method for the infrared temperature monitoring module to obtain the temperature distribution uniformity is as follows: The infrared temperature monitoring module acquires the temperature at each location within the temperature zone, calculates the absolute value of the temperature difference between any two adjacent locations to obtain the temperature difference distribution, and calculates the mean of the temperature difference distribution to obtain the average temperature difference. The infrared temperature monitoring module calculates the variance of all temperatures within the temperature zone to obtain the temperature variance distribution. The infrared temperature monitoring module performs a weighted summation of the average temperature difference and the temperature variance distribution to obtain the temperature distribution uniformity.

[0010] As a preferred embodiment of the present invention, the method for the material dynamic distribution module to generate concrete temperature distribution characteristics and temperature change trend data is as follows: The material dynamic distribution module acquires time-series temperature data, calculates the rate of change of the average temperature in the same temperature zone at adjacent times in the time-series temperature data, obtains the temperature change rate, and performs threshold analysis on the temperature change rate to determine abnormal temperature changes. The material dynamic distribution module calculates the temperature difference between different temperature zones at the same time. If the temperature difference between different temperature zones is greater than a set threshold, it is determined that there is uneven heat of hydration. The material dynamic distribution module feeds back the judgment results of abnormal temperature changes and uneven hydration heat to the main data stream.

[0011] In a preferred embodiment of the present invention, the material dynamic distribution module tracks and analyzes the average temperature change trend of each temperature zone: an average temperature change curve is generated by the average temperature change trend of each temperature zone, and then the average temperature change curves of different temperature zones are fitted to generate a total average temperature change curve. The initial setting time window of concrete is predicted by tracking and analyzing the total average temperature change curve, and the initial setting time window of concrete is published to the main data stream.

[0012] In a preferred embodiment of the present invention, the method for the visual dynamic analysis module to perform dynamic analysis is as follows: S1: Extract multiple consecutive frames of concrete images from the main data stream; S2: Extract the degree of concrete aggregate exposure and the area of ​​surface water bleeding from each frame of image; S3: Calculate the ratios of the degree of concrete aggregate exposure and the area of ​​surface bleeding to the corresponding allowable values ​​to obtain the aggregate exposure characteristic assessment results and the surface bleeding degree assessment results.

[0013] In a preferred embodiment of the present invention, when the integrated early warning module obtains the judgment results of abnormal temperature changes and uneven hydration heat, it generates an early warning of uneven distribution. After obtaining the initial setting time window of the concrete, the comprehensive early warning module calculates the shortest time difference between the initial setting time window and the current time to obtain the remaining time, and then performs a threshold judgment on the remaining time to generate an initial setting early warning. The comprehensive early warning module compares the aggregate exposure characteristic assessment results with the set threshold, and generates an aggregate exposure early warning based on the comparison results. The integrated early warning module compares the surface water seepage assessment results with a set threshold and generates a water seepage warning based on the comparison results.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention generates a common master data stream for visual acquisition and infrared temperature monitoring, and performs parallel analysis on the same master data stream from the temperature dimension and the image dimension respectively. This makes the assessment of concrete transportation quality no longer dependent on a single sensing method. Visual information reflects the apparent state, and temperature information reflects the internal hydration heat and uniformity state. The two are integrated in the master data stream, providing a multi-source complementary judgment basis for comprehensive early warning, and overcoming the defects of incomplete information and easy misjudgment of single monitoring methods.

[0015] 2. This invention also divides the concrete inside the tank into zones, calculates the temperature parameters of each temperature zone, and further calculates the rate of temperature change, determines the uneven heat of hydration, and predicts the initial setting time window. The visual dynamic analysis module extracts the degree of aggregate exposure and the area of ​​surface bleeding area, and calculates the ratio with the allowable value, so that the quality assessment can be quantitatively graded. Based on various triggering conditions such as temperature anomalies, initial setting risk, aggregate exposure, and surface bleeding, it generates warnings for uneven distribution, initial setting, aggregate exposure, and bleeding, so as to achieve multi-level and multi-type refined warning capabilities.

[0016] 3. In this invention, the image feature sequence is used as the main data stream, and the temperature partition and time-series temperature data are merged into the same main data stream. The material dynamic analysis and visual dynamic analysis then obtain the required data from the main data stream respectively, and merge their respective analysis results back into the main data stream. This ensures the uniformity and scalability of the data interfaces between modules, avoids tight coupling and dependence between modules, makes the system easy to maintain and upgrade, and facilitates its deployment on different models of concrete mixer trucks. Attached Figure Description

[0017] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0018] Figure 1 This is a system block diagram of the present invention; Figure 2 This is a system flowchart of the present invention. Detailed Implementation

[0019] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1: Please refer to Figure 1 - Figure 2 As shown, the vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality includes a vision acquisition module, an infrared temperature monitoring module, a material dynamic analysis module, a visual dynamic analysis module, and a comprehensive early warning module. The visual acquisition module acquires images of the concrete inside the concrete mixer truck using an industrial camera deployed above the unloading port or observation tube. The visual acquisition module performs the following preprocessing on the original image: Converting RGB images to grayscale images reduces the computational load of subsequent processing. A 3×3 window median filter is used to denoise the grayscale image, eliminating salt-and-pepper noise and camera sensor noise. Histogram equalization is performed to enhance image contrast and make the boundary between concrete aggregate and grout clearer.

[0021] After preprocessing, the visual acquisition module extracts the following image features from the image: Concrete aggregate exposure level: refers to the ratio of the pixel area of ​​aggregate exposed on the concrete surface to the total pixel area of ​​the image. The specific extraction method is as follows: the edge detection algorithm is used to extract the edge contour of the aggregate, the pixel filling statistics inside the closed contour are performed to obtain the exposed pixel area of ​​the aggregate, and the aggregate exposure level is obtained by dividing it by the total pixel area of ​​the image. Surface water bleeding area: refers to the smooth, reflective area formed on the concrete surface due to the precipitation of free water. The extraction method is as follows: perform adaptive threshold segmentation on the grayscale image, extract the area with high grayscale value and low texture gradient, and count the pixel area of ​​the area to obtain the surface water bleeding area. The visual acquisition module adds an acquisition timestamp and tanker ID to the above image features to form an image feature sequence, which is then sent to subsequent modules as the main data stream. The infrared temperature monitoring module acquires thermal radiation images of the concrete surface inside the tanker using an infrared thermal imager. The grayscale value of each pixel in the thermal radiation image corresponds to the thermal radiation intensity.

[0022] This module converts the grayscale value of each pixel in the thermal radiation image into the actual temperature value according to the pre-calibrated temperature calibration curve, and generates a temperature matrix. Based on the axial and radial structure of the tanker's mixing drum, the infrared temperature monitoring module divides the concrete area into multiple temperature zones. The infrared temperature monitoring module performs the following calculations for each temperature zone: Average temperature: Sum of the temperature values ​​of all pixels in the partition and divide by the total number of pixels; Maximum temperature: The maximum temperature within this zone; Temperature distribution uniformity: This is used to quantify the uniformity of temperature distribution within the zone. The specific calculation method is as follows: For any two adjacent positions, calculate the absolute value of their temperature difference to obtain the set of differences for all pairs of adjacent positions. Calculate the arithmetic mean of this set to obtain the average temperature difference. Calculate the temperature variance distribution: Calculate the variance of all temperature values ​​within this partition; The uniformity of temperature distribution is calculated by weighted summation of the average temperature difference and temperature variance. The infrared temperature monitoring module calculates the average temperature, maximum temperature, and temperature distribution uniformity for each temperature zone, adds timestamps to generate time-series temperature data, and imports each temperature zone and its corresponding time-series temperature data into the main data stream.

[0023] The materials dynamic analysis module obtains time-series temperature data for each temperature zone from the main data stream, performs temperature analysis and trend prediction, as follows: (1) Judgment of abnormal temperature changes: The material dynamics analysis module calculates the average rate of temperature change between two adjacent time frames within the same temperature zone: If the average temperature change rate exceeds the preset rate threshold, it is determined that there is an abnormal temperature change in the partition, and the determination result is fed back to the main data stream; (2) Determination of uneven heat of hydration: At the same time frame, the material dynamic analysis module calculates the temperature difference between different temperature zones. If the temperature difference between different temperature zones is greater than the set threshold, it is determined that there is uneven heat of hydration, indicating that the internal temperature distribution of the concrete is abnormal, and the determination result is fed back to the main data stream. (3) Prediction of initial setting time window: The material dynamic analysis module tracks the average temperature change trend of each temperature zone, connects the average temperature values ​​at each time point to generate an average temperature change curve, and uses a curve fitting method to fit the average temperature change curves of each zone into a total average temperature change curve for multiple temperature zones. Based on the properties of concrete materials, when the overall average temperature change curve reaches an inflection point and the temperature begins to slowly decrease thereafter, the time point when the concrete reaches the initial setting state is predicted by combining the concrete mix proportion and the empirical model of ambient temperature. The time length from the current time to that time point is used as the initial setting time window of the concrete, and the predicted initial setting time window is published to the main data stream.

[0024] The visual dynamic analysis module extracts multiple consecutive frames of concrete images from the main data stream for dynamic analysis and condition assessment. The specific steps are as follows: S1: Extract multiple consecutive frames of concrete images from the main data stream. Each frame of the image has been preprocessed by grayscale conversion, noise reduction, and equalization. S2: Extract the degree of concrete aggregate exposure and the area of ​​surface water bleeding from each frame of image; S3: Calculate the ratio of the above two indicators to their corresponding allowable values ​​to obtain the evaluation results. Specifically, the aggregate exposure characteristic evaluation result is the ratio of the aggregate exposure degree to the set allowable aggregate exposure value. The allowable aggregate exposure value is a threshold preset according to the concrete mix proportion and construction specifications. The surface bleeding degree evaluation result is the ratio of the surface bleeding area to the allowable bleeding area value. The allowable bleeding area value is a threshold set according to the image resolution and engineering experience. The visual dynamics analysis module imports the two evaluation results mentioned above into the main data stream.

[0025] The integrated early warning module obtains temperature analysis results from the material dynamic analysis module and state assessment results from the visual dynamic analysis module from the main data stream. These results include abnormal temperature fluctuations, uneven hydration heat, concrete initial setting time window, aggregate exposure characteristic assessment results, and surface bleeding degree assessment results. Based on the following rules, it generates multi-level early warning signals: Uneven distribution warning: When abnormal temperature change judgment results or uneven hydration heat judgment results are obtained, an uneven distribution warning is generated, indicating that the internal temperature distribution of concrete is abnormal, which may affect the uniformity of quality.

[0026] Initial setting warning: After obtaining the initial setting time window of concrete, calculate the shortest time difference between the initial setting time window and the current time to obtain the remaining time. Then, compare the remaining time with the preset multi-level thresholds: if the remaining time is less than or equal to the first-level threshold, generate a first-level initial setting warning; if the remaining time is less than or equal to the second-level threshold, generate a second-level initial setting warning. The first-level threshold time is longer than the second-level threshold time.

[0027] Aggregate exposure warning: The aggregate exposure characteristic assessment results are compared with the set threshold. If the aggregate exposure characteristic assessment results are greater than the set threshold, an aggregate exposure warning is generated, indicating that too much stone is exposed on the concrete surface, which may be due to segregation caused by over-mixing or mortar loss.

[0028] Bleeding warning: The surface bleeding degree assessment result is compared with the set threshold. If the surface bleeding degree assessment result is greater than or equal to the set threshold, a bleeding warning is generated, indicating that there is too much free water on the concrete surface, which may affect the strength and durability.

[0029] The integrated early warning module pushes the generated early warning signals to the on-site display screen, driver terminal, or remote dispatch platform to realize graded alarm and real-time intervention of transportation quality.

[0030] Thresholds, preset values, or preset ranges are set for result comparison and analysis to determine whether they are good or bad. The magnitude of these values ​​is determined by a combination of large-scale model analysis of the sample data and human experience. They can also be adjusted appropriately based on seasonal or common-sense influence conditions. Similarly, the weighting ratio coefficients and influence factors are set based on the magnitude of each parameter's influence on the results. These values ​​are assigned to reflect the overall impact on the results. They are also determined by a combination of large-scale model analysis of the sample data and human experience. They can also be adjusted appropriately based on seasonal or common-sense influence conditions.

[0031] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality, characterized in that: It includes a visual acquisition module, an infrared temperature monitoring module, a material dynamic analysis module, a visual dynamic analysis module, and a comprehensive early warning module; The visual acquisition module can acquire images of concrete inside the tanker truck through a visual acquisition device, and perform preliminary processing on the acquired concrete images to generate a main data stream. The infrared temperature monitoring module acquires infrared images of the concrete inside the tanker truck using infrared devices, converts the infrared acquisition results into temperature data, divides the concrete inside the tanker truck into regions, generates concrete temperature zones based on the region division results, and imports the temperature zones and corresponding temperature data into the main data stream. The material dynamic analysis module acquires the main data stream, performs temperature analysis based on the main data stream, generates data on concrete temperature distribution characteristics and temperature change trends, and imports the analysis results into the main data stream. The visual dynamic analysis module acquires the main data stream, performs dynamic analysis based on the concrete images, obtains the concrete condition assessment results, and imports the concrete assessment results into the main data stream. The integrated early warning module acquires concrete condition assessment results and temperature analysis results. When any result exceeds the corresponding preset threshold, it generates a multi-level early warning signal and pushes the early warning signal and related data.

2. The vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality according to claim 1, characterized in that, The method for preprocessing the concrete image after acquiring it by the visual acquisition module includes: The original image is preprocessed by grayscale conversion, median filtering for noise reduction, and histogram equalization. The preprocessed images are analyzed to obtain image features including the degree of concrete aggregate exposure and surface water bleeding areas. Add timestamps to image features to form an image feature sequence as the main data stream.

3. The vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality according to claim 1, characterized in that, The infrared temperature monitoring module converts the acquired thermal radiation images into actual temperature values ​​through a calibration curve, generating a temperature matrix. The concrete area is divided according to the mixing drum of the tanker truck, resulting in multiple temperature zones; The infrared temperature monitoring module calculates the average temperature, maximum temperature, and temperature distribution uniformity of each temperature zone to obtain temperature data, and adds a timestamp to the temperature data to generate time-series temperature data. Each temperature zone and the corresponding time-series temperature data are imported into the main data stream.

4. The vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality according to claim 1, characterized in that, The method by which the infrared temperature monitoring module obtains the temperature distribution uniformity is as follows: The infrared temperature monitoring module acquires the temperature at each location within the temperature zone, calculates the absolute value of the temperature difference between any two adjacent locations to obtain the temperature difference distribution, and calculates the mean of the temperature difference distribution to obtain the average temperature difference. The infrared temperature monitoring module calculates the variance of all temperatures within the temperature zone to obtain the temperature variance distribution. The infrared temperature monitoring module performs a weighted summation of the average temperature difference and the temperature variance distribution to obtain the temperature distribution uniformity.

5. The vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality according to claim 1, characterized in that, The method by which the material dynamic distribution module generates concrete temperature distribution characteristics and temperature change trend data is as follows: The material dynamic distribution module acquires time-series temperature data, calculates the rate of change of the average temperature in the same temperature zone at adjacent times in the time-series temperature data, obtains the temperature change rate, and performs threshold analysis on the temperature change rate to determine abnormal temperature changes. The material dynamic distribution module calculates the temperature difference between different temperature zones at the same time. If the temperature difference between different temperature zones is greater than a set threshold, it is determined that there is uneven heat of hydration. The material dynamic distribution module feeds back the judgment results of abnormal temperature changes and uneven hydration heat to the main data stream.

6. The vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality according to claim 1, characterized in that, The material dynamic distribution module tracks and analyzes the average temperature change trend of each temperature zone: it generates an average temperature change curve based on the average temperature change trend of each temperature zone, and then fits the average temperature change curves of different temperature zones to generate an overall average temperature change curve. The initial setting time window of concrete is predicted by tracking and analyzing the total average temperature change curve, and the initial setting time window of concrete is published to the main data stream.

7. The vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality according to claim 1, characterized in that, The method by which the visual dynamic analysis module performs dynamic analysis is as follows: S1: Extract multiple consecutive frames of concrete images from the main data stream; S2: Extract the degree of concrete aggregate exposure and the area of ​​surface water bleeding from each frame of image; S3: Calculate the ratios of the degree of concrete aggregate exposure and the area of ​​surface bleeding to the corresponding allowable values ​​to obtain the aggregate exposure characteristic assessment results and the surface bleeding degree assessment results.

8. The vision-driven multi-source perception and assessment system for concrete mixer truck transportation quality according to claim 1, characterized in that, When the integrated early warning module obtains the judgment results of abnormal temperature changes and uneven hydration heat, it generates an early warning of uneven distribution. After obtaining the initial setting time window of the concrete, the comprehensive early warning module calculates the shortest time difference between the initial setting time window and the current time to obtain the remaining time, and then performs a threshold judgment on the remaining time to generate an initial setting early warning. The comprehensive early warning module compares the aggregate exposure characteristic assessment results with the set threshold, and generates an aggregate exposure early warning based on the comparison results. The integrated early warning module compares the surface water seepage assessment results with a set threshold and generates a water seepage warning based on the comparison results.

Images