Method, system and device for detecting internal quality of red ginseng based on infrared thermal imaging

Abstract

The invention discloses a red ginseng internal quality detection method, system and device based on infrared thermal imaging, and belongs to the field of image processing, and the method comprises the following steps: S1, a sample pretreatment step: pretreating red ginseng to be detected, including cleaning the surface of the red ginseng and fixing the position, and ensuring that the exterior of a sample is clean and the posture is standard; s2, a controlled heating step: performing short-time high-energy heating treatment on the red ginseng by adopting a pulse heating device, and monitoring and adjusting heating parameters in real time by utilizing a closed-loop control system in the heating process to ensure that the internal temperature of the red ginseng reaches local temperature difference and avoid heat balance at the same time; according to the method, the infrared thermal imaging and multispectral fusion technology is utilized, and on the premise that the integrity of the red ginseng is guaranteed, high-precision positioning and recognition of defects such as hollowness, breakage or foreign matter doping in the red ginseng are achieved through fine extraction of the temperature gradient and the local abnormal area.

Description

Technical Field

[0001] The present invention relates to the field of image detection, and more specifically, to a method, system and device for detecting the internal quality of red ginseng based on infrared thermal imaging. Background Art

[0002] As a traditional Chinese medicine with high medicinal value and economic benefits, the quality inspection of red ginseng is crucial to ensure product safety and efficacy. At present, the quality inspection of red ginseng mainly relies on appearance evaluation and chemical composition analysis, but these methods have the following limitations: Traditional quality inspection methods often require physical cutting or chemical treatment of red ginseng, which not only destroys the integrity of the sample, but also may introduce chemical residues or change the internal components of red ginseng, and cannot meet the requirements of non-destructive testing. The manual evaluation method based on characteristics such as appearance and texture is greatly affected by the experience and subjective factors of the inspectors, and it is difficult to ensure the consistency and objectivity of the test results, especially when detecting small defects such as internal hollowness, fracture or foreign matter. Although the existing X-ray, ultrasonic and other detection technologies can partially reflect the internal structure of red ginseng, it is often difficult to distinguish small internal defects due to the small density difference and complex internal structure of red ginseng, and it may involve expensive equipment and complex operations, which limits its application in actual production. Summary of the invention

[0003] The purpose of the present invention is to provide a method, system and device for detecting the internal quality of red ginseng based on infrared thermal imaging, so as to solve the problems raised in the above-mentioned background technology: the traditional quality detection method often requires physical cutting or chemical treatment of red ginseng, which not only destroys the integrity of the sample, but also may introduce chemical residues or change the internal components of red ginseng, and cannot achieve the requirements of non-destructive testing. The manual evaluation method based on the characteristics of appearance, texture, etc. is greatly affected by the experience and subjective factors of the detection personnel, and it is difficult to ensure the consistency and objectivity of the detection results, especially when detecting small defects such as internal hollowness, fracture or foreign matter. Although the existing X-ray, ultrasonic and other detection technologies can partially reflect the internal structure of red ginseng, it is often difficult to distinguish small internal defects due to the small density difference and complex internal structure of red ginseng, and it may involve expensive equipment and complex operations, which limits its application in actual production.

[0004] Technical solution: The internal quality detection method of red ginseng based on infrared thermal imaging includes the following steps: S1. Sample pretreatment step: pre-treat the red ginseng to be tested, including cleaning the surface of the red ginseng and fixing the position to ensure that the sample is clean and in a standard posture; S2, controlled heating step: using a pulse heating device to subject the red ginseng to short-term high-energy heat treatment, and using a closed-loop control system to monitor and adjust the heating parameters in real time during the heating process to ensure that the internal temperature of the red ginseng achieves a local temperature difference while avoiding thermal equilibrium; S3, multispectral image acquisition step: synchronously collect multispectral image data of red ginseng during the heating process, including infrared and visible light images, and ensure spatial correspondence of each image data through optical calibration; S4, image data processing and defect determination steps: pre-process, temperature calibrate and image fuse the collected multi-spectral image data, use image analysis algorithm to extract temperature gradient and local abnormal area, and then identify temperature anomalies caused by internal hollowness, fracture or foreign matter doping.

[0005] Preferably, the S1 further comprises: S1.1. Place the red ginseng to be tested in a testing room with a temperature controlled at 20°C ± 2°C and a relative humidity controlled at 40% to 60%; S1.2. Physically fix the red ginseng to ensure that the position and posture of the sample are consistent during heating and imaging.

[0006] Preferably, S2 adopts a pulse infrared heating method, with a pulse duration of 0.5 to 2 seconds and a pulse interval of 1 to 3 seconds.

[0007] Preferably, the S3 uses an infrared camera and a visible light camera with at least two wavebands, the infrared camera wavebands include short-wave infrared 1-3 μm and long-wave infrared 8-14 μm, the camera acquisition frame rate is not less than 30 Hz, and image synchronous acquisition is achieved through pre-calibration.

[0008] Preferably, the image data processing and defect determination steps include: S4.1, denoising, background correction and temperature calibration of the image; S4.2, identify temperature gradient changes and abnormal areas; S4.3. Fit the temperature diffusion curve of the images at different heating time points, and distinguish normal samples from samples with internal defects based on the temperature gradient characteristics. The standards are as follows: After the internal hollow is heated, the temperature is 3-5°C lower than the surrounding homogeneous tissue. Its temperature gradient characteristics are as follows: at the boundary between the hollow and the surrounding solid tissue, there will be a more obvious temperature gradient jump, and the gradient value will rise sharply; the internal fracture area is interrupted due to the interruption of heat conduction, and its local temperature is 1-3°C lower than the normal area. The temperature difference is not as obvious as the hollow area, and the temperature gradient characteristics of the fracture edge show irregular and locally steep temperature gradients; foreign materials have different thermal conductivity characteristics from red ginseng. If the thermal conductivity of the foreign body is high, the temperature of its area may be higher than the surrounding area; if the thermal conductivity is low, the temperature is low. The temperature difference can generally be between ±2 and 4°C, and the boundary of the foreign body area is clear. The temperature gradient shows a local sudden change, which matches the irregular shape of the red ginseng tissue.

[0009] Red ginseng internal quality inspection system based on infrared thermal imaging, the system includes: A pulse heating unit, wherein the pulse heating unit comprises a high-frequency pulse infrared source, and an energy regulation module and a temperature feedback module thereof constitute a closed-loop control system for short-term and uniform heating of the red ginseng to be tested; A multispectral imaging unit, the unit comprising a short-wave infrared camera, a long-wave infrared camera and a visible light camera, the cameras being spatially aligned optically or electronically to ensure synchronous acquisition of multi-band image data; An intelligent analysis unit, comprising an image preprocessing module, an image segmentation and anomaly detection module based on a convolutional neural network, and a temperature diffusion model analysis module, is used to process and determine internal defects of red ginseng.

[0010] Preferably, the heating parameters of the pulse heating unit are dynamically adjusted by a central processing unit according to real-time temperature monitoring data, so as to fully stimulate the internal temperature difference information before the sample reaches thermal equilibrium.

[0011] Preferably, the multispectral imaging unit achieves an image acquisition rate of at least 30 Hz during the heating process, and converts the multi-band data into a temperature distribution map in a unified coordinate system through an image fusion algorithm for subsequent analysis.

[0012] A red ginseng internal quality detection device based on infrared thermal imaging, the device comprising: (1) a pulse heating module, which is used to achieve high-energy short-time heating of the red ginseng to be tested, and has an internal integrated closed-loop control circuit to automatically adjust the heating parameters according to real-time temperature feedback; (2) Multispectral imaging module, including short-wave infrared, long-wave infrared and visible light imaging devices, with a pre-calibrated optical alignment system to achieve multi-angle and synchronous acquisition; (3) AI image analysis module, based on pre-trained convolutional neural network and temperature diffusion time series algorithm, performs real-time defect recognition and abnormal area location on collected images.

[0013] Compared with the traditional solution, this solution has the following beneficial effects: (1) The present invention utilizes infrared thermal imaging and multi-spectral fusion technology to achieve high-precision positioning and identification of defects such as hollowness, fracture or foreign matter doping inside red ginseng by finely extracting temperature gradients and local abnormal areas while ensuring the integrity of red ginseng.

[0014] (2) The introduction of an image processing algorithm based on a convolutional neural network and a temperature diffusion model can automatically analyze multimodal image data, reduce the interference of human subjective factors, improve the objectivity and stability of the detection results, and facilitate large-scale application.

[0015] (3) Pulse heating and closed-loop control technology are used to quickly collect multi-angle and multi-band data in a controlled environment, and a three-dimensional temperature distribution model is constructed, making real-time online detection possible, thereby effectively improving the level of production quality supervision and market supervision. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic diagram of the appearance of sample 1 of the present invention; Figure 2 Infrared and visible light images of sample 1 of the present invention; Figure 3 This is a schematic diagram of the appearance of sample 2 of the present invention; Figure 4 Infrared and visible light images of sample 2 of the present invention; Figure 5 This is a schematic diagram of the appearance of sample 3 of the present invention; Figure 6 Infrared and visible light images of sample 3 of the present invention; Figure 7 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0017] Example Example 1, please refer to Figure 7 The method for detecting the internal quality of red ginseng based on infrared thermal imaging comprises the following steps: S1. Sample pretreatment step: pre-treat the red ginseng to be tested, including cleaning the surface of the red ginseng and fixing the position to ensure that the sample is clean and in a standard posture; S2, controlled heating step: using a pulse heating device to subject the red ginseng to short-term high-energy heat treatment, and using a closed-loop control system to monitor and adjust the heating parameters in real time during the heating process to ensure that the internal temperature of the red ginseng achieves a local temperature difference while avoiding thermal equilibrium; S3, multispectral image acquisition step: synchronously collect multispectral image data of red ginseng during the heating process, including infrared and visible light images, and ensure spatial correspondence of each image data through optical calibration; S4, image data processing and defect determination steps: pre-process, temperature calibrate and image fuse the collected multi-spectral image data, use image analysis algorithm to extract temperature gradient and local abnormal area, and then identify temperature anomalies caused by internal hollowness, fracture or foreign matter doping.

[0018] Said S1 further comprises: S1.1. Place the red ginseng to be tested in a testing room with a temperature controlled at 20°C ± 2°C and a relative humidity controlled at 40% to 60%; S1.2. Physically fix the red ginseng to ensure that the position and posture of the sample are consistent during heating and imaging.

[0019] The S2 adopts a pulse infrared heating method, with a pulse duration of 0.5 to 2 seconds and a pulse interval of 1 to 3 seconds.

[0020] The S3 uses an infrared camera and a visible light camera with at least two wavebands. The infrared camera wavebands include short-wave infrared 1-3 μm and long-wave infrared 8-14 μm. The camera acquisition frame rate is not less than 30 Hz, and image synchronization acquisition is achieved through pre-calibration.

[0021] The image data processing and defect determination steps include: S4.1, denoising, background correction and temperature calibration of the image; S4.2, identify temperature gradient changes and abnormal areas; S4.3. Fit the temperature diffusion curve of the images at different heating time points, and distinguish normal samples from samples with internal defects based on the temperature gradient characteristics. The standards are as follows: After the internal hollow is heated, the temperature is 3-5°C lower than the surrounding homogeneous tissue. Its temperature gradient characteristics are as follows: at the boundary between the hollow and the surrounding solid tissue, there will be a more obvious temperature gradient jump, and the gradient value will rise sharply; the internal fracture area is interrupted due to the interruption of heat conduction, and its local temperature is 1-3°C lower than the normal area. The temperature difference is not as obvious as the hollow area, and the temperature gradient characteristics of the fracture edge show irregular and locally steep temperature gradients; foreign materials have different thermal conductivity characteristics from red ginseng. If the thermal conductivity of the foreign body is high, the temperature of its area may be higher than the surrounding area; if the thermal conductivity is low, the temperature is low. The temperature difference can generally be between ±2 and 4°C, and the boundary of the foreign body area is clear. The temperature gradient shows a local sudden change, which matches the irregular shape of the red ginseng tissue.

[0022] Combined with the temperature gradient map, CNN segmentation results and temperature diffusion curve, characteristic indicators such as average temperature difference, gradient mean, gradient variance, etc. are calculated for each candidate area; Threshold determination: Set the threshold for temperature difference and gradient change. For example: If the average temperature of the area is more than 3°C lower than the surrounding area and the gradient mutation value exceeds the preset threshold, it is judged as hollow; If the area shows irregular temperature changes and abnormal local temperature changes, it is judged to be a fracture; If the direction of the regional temperature anomaly is significantly different from the surrounding tissue and the shape boundary is clear, it is judged to be foreign matter. Red ginseng internal quality inspection system based on infrared thermal imaging, the system includes: A pulse heating unit, wherein the pulse heating unit comprises a high-frequency pulse infrared source, and an energy regulation module and a temperature feedback module thereof constitute a closed-loop control system for short-term and uniform heating of the red ginseng to be tested; A multispectral imaging unit, the unit comprising a short-wave infrared camera, a long-wave infrared camera and a visible light camera, the cameras being spatially aligned optically or electronically to ensure synchronous acquisition of multi-band image data; An intelligent analysis unit, comprising an image preprocessing module, an image segmentation and anomaly detection module based on a convolutional neural network, and a temperature diffusion model analysis module, is used to process and determine internal defects of red ginseng.

[0023] The heating parameters of the pulse heating unit are dynamically adjusted by the central processor according to real-time temperature monitoring data, so as to fully stimulate the internal temperature difference information before the sample reaches thermal equilibrium.

[0024] The multi-spectral imaging unit achieves an image acquisition rate of at least 30 Hz during the heating process, and converts the multi-band data into a temperature distribution map in a unified coordinate system through an image fusion algorithm for subsequent analysis.

[0025] A red ginseng internal quality detection device based on infrared thermal imaging, the device comprising: (1) a pulse heating module, which is used to achieve high-energy short-time heating of the red ginseng to be tested, and has an internal integrated closed-loop control circuit to automatically adjust the heating parameters according to real-time temperature feedback; (2) Multispectral imaging module, including short-wave infrared, long-wave infrared and visible light imaging devices, with a pre-calibrated optical alignment system to achieve multi-angle and synchronous acquisition; (3) AI image analysis module, based on pre-trained convolutional neural network and temperature diffusion time series algorithm, performs real-time defect recognition and abnormal area location on collected images.

[0026] Example 2: Please refer to Figure 1-6 , three groups of red ginseng samples were designed for testing.

[0027] Figure 1-2 The sample in the middle is broken and hollow. After the red ginseng is heated, the fracture and hollowness lead to inconsistent heat absorption, and light and shadow areas of different degrees appear. It can be judged that the internal non-uniform distribution of the red ginseng, that is, the fracture structure, is different. Because the density of the hollow and fractured media is different, the heat collection degree is different, resulting in structural change characteristics in the imaging.

[0028] Figure 3-4 The medium sample is qualified. It is heated for a short time. The thermal imaging of the coarse red ginseng shows that the inside is uniform, without obvious light and shadow areas of different degrees. Because the internal medium of the red ginseng is uniform, if it is heated for too long, the infrared image will also be uniform.

[0029] Figure 5-6 The same sample Figure 3-4 The middle sample is also a qualified product.

[0030] Hollow inside Difference in temperature values: Since the hollow area is filled with air, its heat capacity and conductivity are low, and after heating, its temperature is usually 3 to 5°C lower than the surrounding homogeneous tissue.

[0031] Temperature gradient characteristics: At the boundary between the hollow core and the surrounding solid tissue, there will be a more obvious temperature gradient jump, and the gradient value will rise sharply.

[0032] Time response: The temperature response of the hollow area is slow, and there is a significant temperature difference with the normal area in the initial stage of heating, but it tends to be uniform after long-term heating. Therefore, the detection time window should be controlled in the initial stage.

[0033] Internal fracture (crack) Temperature value difference: Due to the interruption of heat conduction, the local temperature of the fractured area may be 1-3°C lower than that of the normal area, but the temperature difference may not be as obvious as that of the hollow area; Temperature gradient characteristics: The fracture edge usually presents irregular and locally steep temperature gradients, and the gradient changes are more discontinuous and dispersed than in the hollow area.

[0034] Time response: The temperature diffusion in the fracture area presents an irregular curve, the local temperature changes rapidly, and the temperature distribution in the crack extension direction shows obvious discontinuity.

[0035] Foreign matter doping Temperature value difference: Foreign materials usually have different thermal conductivity characteristics from red ginseng. If the foreign material has high thermal conductivity, the temperature of the area may be higher than the surrounding area; if the thermal conductivity is low, the temperature is lower. The temperature difference is generally between ±2 and 4℃.

[0036] Temperature gradient characteristics: The boundary of the foreign body area is clear, and the temperature gradient shows a local sudden change, which matches the irregular shape of the red ginseng tissue. Time response: The temperature response of the foreign body area is significantly different from that of the surrounding tissue, and its thermal diffusion curve is obviously inconsistent with that of the red ginseng tissue, showing a stable abnormal state maintained for a long time after a rapid rise or fall in temperature.

[0037] The above shows and describes the basic principles, main features and advantages of the present invention; those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected; the scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for detecting the internal quality of red ginseng based on infrared thermal imaging, characterized in that: The following steps are involved: S1. Sample pretreatment step: pre-treat the red ginseng to be tested, including cleaning the surface of the red ginseng and fixing the position to ensure that the sample is clean and in a standard posture; S2, controlled heating step: using a pulse heating device to subject the red ginseng to short-term high-energy heat treatment, and using a closed-loop control system to monitor and adjust the heating parameters in real time during the heating process to ensure that the internal temperature of the red ginseng achieves a local temperature difference while avoiding thermal equilibrium; S3, multispectral image acquisition step: synchronously acquire infrared and visible light images of red ginseng during the heating process, and ensure spatial correspondence of image data through optical calibration; S4, image data processing and defect determination steps: pre-process, temperature calibrate and image fuse the collected infrared and visible light images, use image analysis algorithms to extract temperature gradients and local abnormal areas, and then identify temperature anomalies caused by internal hollowness, fractures or foreign matter doping.

2. The method for detecting the internal quality of red ginseng based on infrared thermal imaging according to claim 1, characterized in that: The S1 further comprises: S1.

1. Place the red ginseng to be tested in a testing room with a temperature controlled at 20°C ± 2°C and a relative humidity controlled at 40% to 60%; S1.

2. Physically fix the red ginseng to ensure that the position and posture of the sample are consistent during heating and imaging.

3. The method for detecting the internal quality of red ginseng based on infrared thermal imaging according to claim 1, characterized in that: The S2 adopts a pulse infrared heating method, with a pulse duration of 0.5 to 2 seconds and a pulse interval of 1 to 3 seconds.

4. The method for detecting the internal quality of red ginseng based on infrared thermal imaging according to claim 1, characterized in that: The S3 uses an infrared camera and a visible light camera with at least two wavebands. The infrared camera wavebands include short-wave infrared 1-3 μm and long-wave infrared 8-14 μm. The camera acquisition frame rate is not less than 30 Hz, and image synchronization acquisition is achieved through pre-calibration.

5. The method for detecting the internal quality of red ginseng based on infrared thermal imaging according to claim 1, characterized in that: The image data processing and defect determination steps include: S4.1, denoising, background correction and temperature calibration of the image; S4.2, identify temperature gradient changes and abnormal areas; S4.

3. Fit the temperature diffusion curve of the images at different heating time points, and distinguish normal samples from samples with internal defects based on the temperature gradient characteristics. The standards are as follows: After the internal hollow is heated, the temperature is 3-5°C lower than the surrounding homogeneous tissue. Its temperature gradient characteristics are as follows: at the boundary between the hollow and the surrounding solid tissue, there will be a more obvious temperature gradient jump, and the gradient value will rise sharply; the internal fracture area is interrupted due to the interruption of heat conduction, and its local temperature is 1-3°C lower than the normal area. The temperature difference is not as obvious as the hollow area, and the temperature gradient characteristics of the fracture edge show irregular and locally steep temperature gradients; foreign materials have different thermal conductivity characteristics from red ginseng. If the thermal conductivity of the foreign body is high, the temperature of its area may be higher than the surrounding area; if the thermal conductivity is low, the temperature is low. The temperature difference can generally be between ±2 and 4°C, and the boundary of the foreign body area is clear. The temperature gradient shows a local sudden change, which matches the irregular shape of the red ginseng tissue.

6. The red ginseng internal quality detection system based on infrared thermal imaging is characterized by: The system includes: A pulse heating unit, wherein the pulse heating unit comprises a high-frequency pulse infrared source, and an energy regulation module and a temperature feedback module thereof constitute a closed-loop control system for short-term and uniform heating of the red ginseng to be tested; A multispectral imaging unit, the unit comprising a short-wave infrared camera, a long-wave infrared camera and a visible light camera, the cameras being spatially aligned optically or electronically to ensure synchronous acquisition of multi-band image data; An intelligent analysis unit, comprising an image preprocessing module, an image segmentation and anomaly detection module based on a convolutional neural network, and a temperature diffusion model analysis module, is used to process and determine internal defects of red ginseng.

7. The red ginseng internal quality detection system based on infrared thermal imaging according to claim 6, characterized in that: The heating parameters of the pulse heating unit are dynamically adjusted by the central processor according to real-time temperature monitoring data, so as to fully stimulate the internal temperature difference information before the sample reaches thermal equilibrium.

8. The red ginseng internal quality detection system based on infrared thermal imaging according to claim 6, characterized in that: The multi-spectral imaging unit achieves an image acquisition rate of at least 30 Hz during the heating process, and converts the multi-band data into a temperature distribution map in a unified coordinate system through an image fusion algorithm for subsequent analysis.

9. A red ginseng internal quality detection device based on infrared thermal imaging, characterized in that: The device includes: (1) a pulse heating module, which is used to achieve high-energy short-time heating of the red ginseng to be tested, and has an internal integrated closed-loop control circuit to automatically adjust the heating parameters according to real-time temperature feedback; (2) Multispectral imaging module, including short-wave infrared, long-wave infrared and visible light imaging devices, with a pre-calibrated optical alignment system to achieve multi-angle and synchronous acquisition; (3) AI image analysis module, based on pre-trained convolutional neural network and temperature diffusion time series algorithm, performs real-time defect recognition and abnormal area location on collected images.

Images