Camellia oleifera grafting quality evaluation method and system

Abstract

The embodiment of the invention provides a camellia oleifera grafting quality assessment method and system, and relates to the technical field of camellia oleifera grafting quality assessment, and the method comprises the steps: collecting a hyperspectral spectrum of a graft union region under a preset wave band, and obtaining the hyperspectral spectrum of the graft union region; obtaining a moisture index of the graft union area according to the hyperspectral spectrum of the graft union area; preset temperature stimulation is applied to the graft union area, and a physiological response infrared image sequence of the graft union area after temperature stimulation is obtained; calculating the temperature recovery rate of the graft union area according to the infrared image sequence; evaluating the temperature distribution uniformity according to the infrared image sequence; and judging the healing quality of the graft union according to the moisture index, the temperature recovery rate or the temperature distribution uniformity. The survival rate and the subsequent growth quality of the camellia oleifera grafted seedlings can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of camellia oleifera grafting quality evaluation, in particular to a camellia oleifera grafting quality evaluation method and system. BACKGROUND

[0002] In related technologies, the camellia oleifera grafting quality evaluation system, in order to improve efficiency and standardization in large-scale planting of camellia oleifera, people introduced a grafting quality evaluation system based on computer vision. This kind of system is originally designed to quickly distinguish between "obvious survival" and "obvious failure" of grafting interface. However, as the source of grafting materials becomes diverse, a phenomenon called "false healing" becomes more and more common. This kind of grafting interface looks very healthy from the outside, but in fact the internal channel responsible for transporting water and nutrients is not fully connected or not firmly connected. The existing evaluation system is difficult to find these internal problem seedlings because it only looks at the surface phenomenon, resulting in them being incorrectly treated as healthy seedlings, and ultimately experiencing unexpected survival rate decline in subsequent growth. SUMMARY

[0003] The present application aims to at least solve one of the technical problems existing in the prior art. To this end, the present application provides a camellia oleifera grafting quality evaluation method and system, which aims to accurately distinguish between healthy healing and false healing, avoid false healing seedlings being incorrectly treated, and effectively improve the survival rate and subsequent growth quality of camellia oleifera grafting seedlings.

[0004] In a first aspect, the embodiments of the present application provide a camellia oleifera grafting quality evaluation method, comprising: Collecting a hyperspectral spectrum of the grafting interface area under a preset waveband to obtain a hyperspectral spectrum of the grafting interface area; Obtaining a water index of the grafting interface area according to the hyperspectral spectrum of the grafting interface area; Applying a preset temperature stimulus to the grafting interface area to obtain a physiological response infrared image sequence of the grafting interface area after temperature stimulation; Calculating a temperature recovery rate of the grafting interface area according to the infrared image sequence; Evaluating temperature distribution uniformity according to the infrared image sequence; Judging the healing quality of the grafting interface according to the water index, the temperature recovery rate or the temperature distribution uniformity.

[0005] Further, in the step of obtaining the water index of the grafting interface area, specifically comprising: Extracting the reflection intensity R_1300 at a wavelength of 1300 nm and the reflection intensity R_1450 at a wavelength of 1450 nm from the hyperspectral spectrum of the grafting interface area; Based on the reflection intensities R_1300 and R_1450, the moisture index of the grafting interface area is calculated as follows: Moisture index of the grafting interface area = (R_1300 - R_1450) / (R_1300 + R_1450).

[0006] In some preferred embodiments, the step of obtaining the moisture index of the grafting interface region specifically includes: Obtain the hyperspectral spectrum of the healthy diameter segment within a certain range of the grafting interface area, where the outer perimeter is a region approximately 10 cm from the grafting interface area. The hyperspectral spectrum of the grafting interface region was corrected based on the hyperspectral spectrum of the healthy segment to obtain the corrected hyperspectral spectrum of the grafting interface region. The moisture index of the grafting interface region was obtained based on the corrected hyperspectral image of the grafting interface region.

[0007] Furthermore, the step of calculating the temperature recovery rate of the grafting interface region specifically includes: Each frame of the infrared image sequence is divided into sub-regions to form multiple sub-region infrared image sequences. The temperature recovery rate of each sub-region is calculated based on multiple sub-region infrared image sequences. Based on the temperature recovery rate of each sub-region and its corresponding area weight, a weighted average temperature recovery rate is calculated and used as the temperature recovery rate of the grafting interface region.

[0008] Preferably, the step of calculating the temperature recovery rate of each sub-region based on multiple sub-region infrared image sequences specifically includes: Obtain the ambient temperature when a preset temperature stimulus is applied to the grafting interface area; Extract the curve of the average surface temperature of the grafting interface region over time from the infrared image sequence of the sub-region; Based on the curve, the temperature recovery rate of the sub-region is calculated by fitting an exponential recovery model. The temperature recovery rate of the sub-region is calculated according to the following formula: T(t) = T_env - (T_env - T_min) exp(-kt) Where T(t) is the surface temperature at time t, T_env is the ambient temperature, T_min is the lowest temperature that the grafting interface region can reach after applying a preset temperature stimulus, k is the temperature recovery rate, and t is time.

[0009] Based on the above, this application further proposes that the step of evaluating the uniformity of temperature distribution specifically includes: Based on the infrared image sequence, multiple surface temperatures of multiple sub-regions were obtained at a certain time after stimulation; Calculate the standard deviation of surface temperature based on multiple surface temperatures; The standard deviation of surface temperature is used as an index of temperature uniformity to evaluate the uniformity of temperature distribution.

[0010] As a technical improvement, after calculating the moisture index of the grafting interface area based on the reflection intensities R_1300 and R_1450, the method further includes: Obtain ambient humidity; The moisture index of the grafting interface area is adjusted according to the ambient humidity. When the ambient humidity is greater than the preset humidity threshold, the moisture index is lowered; when the ambient humidity is less than the preset humidity threshold, the moisture index is raised.

[0011] To improve the solution, the steps for assessing the healing quality of the graft union specifically include: Obtain information on the variety, growth stage, and environmental humidity of the grafted seedlings; Based on variety information, growth stage information, and environmental humidity information, determine the moisture index of healthy grafting interface, the temperature recovery rate of healthy grafting interface, and the temperature distribution uniformity of healthy grafting interface. The difference between the moisture index and the moisture index of the healthy graft interface is used to obtain the first difference value; the difference between the temperature recovery rate and the temperature recovery rate of the healthy graft interface is used to obtain the second difference value; and the difference between the temperature distribution uniformity and the temperature distribution uniformity of the healthy graft interface is used to obtain the third difference value. The healing quality of the grafting interface is determined based on the first, second, and third differences.

[0012] In specific cases, the steps for determining the healing quality of the graft union based on the first, second, and third differences specifically include: When the first difference is greater than the preset moisture index difference threshold, or the second difference is greater than the temperature recovery rate difference threshold, or the third difference is less than the temperature distribution uniformity threshold, the grafting interface is judged to be a false healing.

[0013] Secondly, this application also discloses a Camellia oleifera grafting quality assessment system for evaluating the graft union healing quality of Camellia oleifera grafted seedlings. The system includes: The hyperspectral image acquisition module is used to collect the hyperspectral image of the grafting interface region in a preset band and obtain the hyperspectral image of the grafting interface region. The moisture index calculation module is used to obtain the moisture index of the grafting interface area based on the hyperspectral image of the grafting interface area. The infrared image sequence acquisition module is used to apply a preset temperature stimulus to the grafting interface area and acquire the infrared image sequence of the physiological response of the grafting interface area after the temperature stimulus. The temperature recovery rate calculation module is used to calculate the temperature recovery rate of the grafting interface area based on the infrared image sequence. The temperature distribution uniformity assessment module is used to assess the temperature distribution uniformity based on the infrared image sequence. The judgment module is used to determine the healing quality of the grafting interface based on the moisture index, temperature recovery rate, or temperature distribution uniformity.

[0014] The method for assessing the grafting quality of Camellia oleifera disclosed in this application obtains the moisture index by collecting hyperspectral images of the grafting interface area, and acquires infrared image sequences of the physiological response after applying temperature stimulation to the grafting interface area to calculate the temperature recovery rate and assess the uniformity of temperature distribution. Finally, the healing quality of the grafting interface is judged by comprehensively considering the moisture index, temperature recovery rate, and temperature distribution uniformity. This method can comprehensively and deeply assess the grafting interface from multiple dimensions such as moisture status, physiological activity, and tissue structure integrity, effectively solving the technical problem of existing technologies that rely solely on surface phenomena and cannot identify "false healing." By introducing hyperspectral and infrared thermal imaging technologies, this application can capture the physiological changes and healing status inside the grafting interface, thereby accurately distinguishing between healthy healing and false healing, avoiding the incorrect handling of seedlings with false healing, and effectively improving the survival rate and subsequent growth quality of grafted Camellia oleifera seedlings, demonstrating significant technological progress and practical value.

[0015] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0016] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0017] Figure 1 This is a flowchart illustrating a method for evaluating the grafting quality of Camellia oleifera according to one embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical methods, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] It should be noted that the meaning of "multiple" (or "more than") in the description of the embodiments of this application refers to two or more, and "greater than," "less than," "exceeding," etc. are understood to exclude the number itself, while "above," "below," "within," etc. are understood to include the number itself. If "first," "second," etc. are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0020] Based on the above, this application proposes a method and system for evaluating the grafting quality of Camellia oleifera, which aims to accurately distinguish between healthy healing and false healing, avoid the incorrect handling of seedlings with false healing, and effectively improve the survival rate and subsequent growth quality of grafted Camellia oleifera seedlings.

[0021] The camellia oleifera grafting quality assessment method provided in this application embodiment can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms; the software can be an application that implements the camellia oleifera grafting quality assessment method, but is not limited to the above forms.

[0022] This application can be applied to numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via communication networks. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices. It should be noted that in various specific embodiments of this invention, when processing is required based on data related to the characteristics of an object (e.g., user attributes or sets of attribute information), permission or consent from the corresponding object is obtained first, and the collection, use, and processing of this data comply with relevant laws and standards. Furthermore, when the embodiments of the present invention need to obtain the attribute information of an object, they will obtain the separate permission or separate consent of the corresponding object through pop-up windows or redirection to a confirmation page. After obtaining the separate permission or separate consent of the corresponding object, they will then obtain the relevant data of the object necessary for the embodiments of the present invention to operate normally.

[0023] See Figure 1 , Figure 1 This is a flowchart illustrating a method for evaluating the grafting quality of Camellia oleifera according to an embodiment of this application. The method for evaluating the grafting quality of Camellia oleifera according to this embodiment includes, but is not limited to, steps S110 to S160, which are described below.

[0024] S110. Collect the hyperspectral spectrum of the grafting interface area in the preset band to obtain the hyperspectral spectrum of the grafting interface area. S120. Obtain the moisture index of the grafting interface area based on the hyperspectral image of the grafting interface area; S130. Apply a preset temperature stimulus to the grafting interface area and obtain an infrared image sequence of the physiological response of the grafting interface area after the temperature stimulus. S140. Calculate the temperature recovery rate of the grafting interface area based on the infrared image sequence; S150. Evaluate the uniformity of temperature distribution based on the infrared image sequence; S160. The healing quality of the grafting interface is judged based on the moisture index, temperature recovery rate, or temperature distribution uniformity.

[0025] To better understand the camellia grafting quality assessment method proposed in this application, it is necessary to explain some of the key terms involved.

[0026] The "grafting interface area" refers to the site where the rootstock and scion unite in a grafted camellia seedling, and is the core area for assessing healing quality. "Hyperspectral data" refers to the reflectance or transmission spectral data of an object collected within multiple continuous narrow bands. It provides richer spectral information than traditional multispectral images, revealing the fine composition and physiological state of the material. "Moisture index" is an indicator calculated based on hyperspectral data, reflecting the water content in plant tissues, and is usually closely related to the plant's health and physiological activity. "Temperature stimulation" refers to applying a preset temperature change, such as heating or cooling, to the grafting interface area to induce a physiological response. "Physiological response infrared image sequence" refers to the continuous acquisition of surface temperature images of the grafting interface area using an infrared thermal imager during temperature stimulation. These images dynamically reflect the temperature changes and heat conduction characteristics of the grafting interface area. "Temperature recovery rate" refers to the speed at which the surface temperature of the grafting interface area recovers to its initial state after being subjected to temperature stimulation; this rate is related to the tissue's metabolic activity and water transport capacity. "Temperature distribution uniformity" refers to the spatial consistency of surface temperature distribution in the grafting area after being stimulated by temperature. Uneven temperature distribution may indicate poor internal healing or tissue damage.

[0027] The method for assessing the grafting quality of Camellia oleifera proposed in this application achieves a comprehensive assessment of the healing quality of the graft union through a series of steps.

[0028] First, it is necessary to acquire hyperspectral images of the graft union region at a preset wavelength to obtain a hyperspectral image of the graft union region. This step can be accomplished using a hyperspectral imager. For example, the hyperspectral imager can be placed above the graft union region, and its acquisition wavelength range can be set, such as the visible to near-infrared band (400nm-1700nm). Then, the device can be started to scan, thereby acquiring the reflectance intensity data of the graft union region at different wavelengths, forming a hyperspectral image.

[0029] Secondly, the moisture index of the grafting interface region is obtained based on the hyperspectral image of the region. For example, specific bands related to moisture absorption, such as the reflection intensity at 1300 nm and 1450 nm, can be selected from the acquired hyperspectral image. By calculating the ratio or difference of the reflection intensity of these bands, the moisture index reflecting the water content of the grafting interface region can be obtained.

[0030] Next, a preset temperature stimulus is applied to the grafting interface area, and an infrared image sequence of the physiological response of the grafting interface area after the temperature stimulus is obtained. For example, a small heater or cooling device can be used to locally heat or cool the grafting interface area for a short period of time, causing its surface temperature to change. At the same time, an infrared thermal imager is used to continuously capture infrared images of the grafting interface area at a certain frame rate (e.g., 10 frames per second), recording the sequence of its surface temperature changes over time.

[0031] Next, the temperature recovery rate of the grafting interface region is calculated based on the infrared image sequence. For example, a curve showing the change of the average surface temperature of the grafting interface region over time can be extracted from the infrared image sequence. By mathematically fitting this curve, such as using an exponential recovery model, the temperature recovery rate of the grafting interface region can be calculated. This rate reflects the heat conduction and metabolic capacity of the grafting interface region.

[0032] Subsequently, the uniformity of temperature distribution is assessed based on the infrared image sequence. For example, multiple frames can be selected from the infrared image sequence within a certain period after stimulation, and the surface temperature of different sub-regions within the grafting interface area can be calculated. Then, by calculating the standard deviation of the surface temperatures of these sub-regions, this standard deviation is used as an index of temperature uniformity, thereby assessing the uniformity of temperature distribution in the grafting interface area.

[0033] Finally, the healing quality of the graft union is assessed based on the moisture index, temperature recovery rate, or temperature distribution uniformity. For example, the calculated moisture index, temperature recovery rate, and temperature distribution uniformity can be compared with preset healthy graft union standard values. If these indicators deviate significantly from the healthy standard values, it can be determined that the graft union has poor healing or pseudo-healing issues.

[0034] Traditional Camellia oleifera grafting quality assessment systems primarily rely on observing the surface morphology of the graft union, such as color and degree of swelling. These external characteristics often fail to accurately reflect the internal healing status of the graft union. When "false healing" occurs, the external appearance of the graft union appears normal, but the internal vascular bundles are not properly connected, leading to obstructed water and nutrient transport and ultimately affecting the survival rate of the grafted seedlings.

[0035] This application utilizes hyperspectral image analysis to obtain the moisture index of the grafting interface region. The moisture index is an important indicator of plant physiological activity; a healthy grafting interface typically exhibits high moisture content and active water transport capacity. Quantitative assessment of the moisture index allows for a preliminary determination of the water supply status within the grafting interface.

[0036] Furthermore, this application introduces infrared image sequence analysis of the physiological response after temperature stimulation. When the graft union region is stimulated by temperature, its internal metabolic activity and water transport capacity affect the speed and distribution of temperature recovery. Healthy graft unions, due to good vascular bundle connections and unimpeded water transport, typically exhibit a faster temperature recovery rate and more uniform temperature distribution. However, in pseudo-healed graft unions, due to poor internal connections and obstructed water transport, the temperature recovery rate may be slower, or localized temperature anomalies may occur, resulting in uneven temperature distribution. By calculating the temperature recovery rate and assessing the uniformity of temperature distribution, the physiological functional state within the graft union can be revealed more deeply.

[0037] Therefore, the method proposed in this application, by comprehensively analyzing three key indicators—moisture index, temperature recovery rate, and temperature distribution uniformity—can comprehensively and objectively reflect the healing quality of the graft union. This multi-dimensional, non-contact evaluation method can not only accurately distinguish between graft unions that are "obviously successful" and those that are "obviously unsuccessful," but more importantly, it can effectively identify the "false healing" phenomenon that is difficult to detect using traditional methods. Compared with existing technologies, the evaluation method of this application has higher accuracy and reliability, can significantly improve the survival rate of grafted Camellia oleifera seedlings, reduce planting risks, and provide strong technical support for the large-scale planting of Camellia oleifera.

[0038] In some embodiments, the step of obtaining the moisture index of the grafting interface region based on the hyperspectral spectrum of the grafting interface region includes: Based on the hyperspectral spectrum of the grafting interface region, the reflection intensity R_1300 at a wavelength of 1300nm and the reflection intensity R_1450 at a wavelength of 1450nm were extracted. Based on the reflection intensities R_1300 and R_1450, the moisture index of the grafting interface area is calculated as follows: Moisture index of the grafting interface area = (R_1300 - R_1450) / (R_1300 + R_1450).

[0039] Specifically, the reflectance intensities R_1300 at 1300 nm and R_1450 at 1450 nm refer to reflectance data obtained from the grafting interface region using hyperspectral imaging technology in specific near-infrared bands. These bands were chosen because they are closely related to the water absorption characteristics in plant tissues. For example, the 1450 nm band is a strong absorption band for water absorption, while the 1300 nm band is a relatively weak reference band. By extracting the reflectance intensities of these two bands, the water content of the grafting interface region can be effectively reflected. The formula for calculating the water index of the grafting interface region, (R_1300 - R_1450) / (R_1300 + R_1450), is a typical form of the Normalized Difference Water Index (NDWI). This formula, by differentiating and normalizing the reflectance intensities of the two bands, aims to eliminate the influence of non-water factors such as light conditions and background reflection on the measurement results, thereby more accurately highlighting the relative water content of the grafting interface region. When the graft union heals well, water transport is usually smooth, and the moisture content in the graft union area is relatively high, resulting in a specific range of moisture index values. Conversely, if healing is poor, water transport is obstructed, the moisture content may decrease, and the moisture index will also change accordingly.

[0040] This application's method effectively quantifies the relative moisture content of the grafting interface region by precisely extracting the reflection intensity at specific wavelengths of 1300nm and 1450nm and calculating it using a normalized difference formula. The essence of plant grafting healing is the formation of new vascular bundle connections between the rootstock and scion to restore water and nutrient transport. Water is fundamental to plant life activities, and the moisture state of the grafting interface region directly reflects the physiological activity of the callus tissue and the connectivity of the vascular bundles. By monitoring the reflection intensity of these two moisture-sensitive wavelengths, subtle changes in water transport efficiency during grafting interface healing can be captured, thus providing direct physiological evidence for assessing healing quality.

[0041] The above technical solution provides a specific, quantitative, and highly interference-resistant method for obtaining the moisture index of the grafting interface area. This method, based on the reflection intensity of a specific wavelength band, effectively eliminates interference from non-moisture factors such as changes in ambient light and background reflection, making the calculated moisture index more stable and reliable. Therefore, it can more accurately reflect the actual moisture status of the grafting interface area, significantly improving the accuracy and sensitivity of the moisture index in assessing the grafting quality of Camellia oleifera, and providing more precise data support for subsequent healing quality judgment.

[0042] Specifically, the steps for obtaining the moisture index of the grafting interface region based on the hyperspectral image of the grafting interface region include: Obtain the hyperspectral spectrum of the healthy diameter segment within a certain range of the grafting interface area, wherein the certain outer perimeter is a region approximately 10 cm from the grafting interface area. The hyperspectral spectrum of the grafting interface region is corrected based on the hyperspectral spectrum of the healthy diameter segment to obtain the corrected hyperspectral spectrum of the grafting interface region. Based on the corrected hyperspectral image of the grafting interface region, the moisture index of the grafting interface region is obtained.

[0043] Obtaining the hyperspectral spectrum of a healthy stem segment within a certain range of the grafting interface area refers to selecting a healthy stem segment unaffected by grafting near the grafting interface area and collecting its hyperspectral spectrum. The phrase "a certain perimeter is an area approximately 10cm above or below the grafting interface area" aims to ensure a high degree of consistency between the selected healthy stem segment and the grafting interface area in terms of physiological and environmental conditions, thus enabling the hyperspectral spectrum of the healthy stem segment to serve as a valid reference benchmark. For example, a healthy stem segment approximately 10cm above or below the grafting interface can be scanned using a hyperspectral imaging device to obtain its hyperspectral data.

[0044] Furthermore, correcting the hyperspectral spectrum of the grafting interface region based on the hyperspectral spectrum of the healthy stem segment refers to calibrating or normalizing the original hyperspectral spectrum of the grafting interface region using the hyperspectral data of the healthy stem segment as a reference. The aim is to eliminate or reduce the influence of non-grafting healing factors (such as changes in ambient light, instrument drift, and fluctuations in the overall plant moisture status) on the hyperspectral spectrum of the grafting interface region, thereby highlighting the spectral changes caused by the grafting interface healing state itself. Correction methods may include, but are not limited to, spectral normalization, baseline correction, and scattering correction.

[0045] Therefore, after obtaining the corrected hyperspectral image of the grafting interface region, the moisture index of the grafting interface region can be obtained based on this corrected image. For example, a more accurate moisture index can be calculated from the corrected hyperspectral data using specific spectral index formulas (such as NDWI, WI, etc.) or machine learning models.

[0046] This application's solution uses the hyperspectral spectrum of a healthy stem segment as a reference to correct the hyperspectral spectrum of the grafting interface area. The working principle is as follows: the healthy stem segment represents the normal spectral response of the grafted Camellia oleifera seedling under current environmental and physiological conditions. When the hyperspectral spectrum of the grafting interface area is affected by environmental or non-healing physiological factors, these effects are usually reflected to some extent in the hyperspectral spectrum of the healthy stem segment. By comparing and correcting the hyperspectral spectrum of the grafting interface area with that of the healthy stem segment, the spectral changes caused by the healing state of the grafting interface itself can be effectively separated, thereby eliminating or reducing the negative impact of other interfering factors on the moisture index calculation. Specifically, obtaining the hyperspectral spectrum of the healthy stem segment provides a baseline of "normal" conditions, while correcting the hyperspectral spectrum of the grafting interface area involves adjusting this baseline so that the corrected spectrum more accurately reflects the local moisture status of the grafting interface area, rather than moisture changes caused by overall or environmental factors.

[0047] By employing the aforementioned technical solution, the hyperspectral image of the grafting interface area is corrected by the hyperspectral image of the healthy diameter segment before obtaining the moisture index. This effectively eliminates the influence of environmental factors and non-healing physiological factors on the spectral data, allowing the calculated moisture index of the grafting interface area to more accurately reflect the true healing state of the graft. Compared to methods that directly obtain the moisture index from the original hyperspectral image, this solution significantly improves the accuracy and reliability of the moisture index, thus providing more precise and stable data support for assessing the grafting quality of Camellia oleifera, reducing the risk of misjudgment, and ultimately improving the efficiency and scientific rigor of grafted seedling selection and management.

[0048] In some preferred embodiments, the specific implementation is as follows: First, a hyperspectral imager is used to scan the grafting point area of ​​a grafted camellia oleifera seedling to obtain its original hyperspectral spectrum. Simultaneously, a healthy stem segment is selected approximately 5 cm above the grafting point, and its hyperspectral spectrum is obtained using the same hyperspectral imager. Then, the original hyperspectral spectrum of the grafting point area and the hyperspectral spectrum of the healthy stem segment are input into a data processing unit. The data processing unit uses a preset correction algorithm (e.g., normalization based on the spectral data of the healthy stem segment, or elimination of background noise and ambient light differences through methods such as difference or ratio) to correct the original hyperspectral spectrum of the grafting point area, obtaining a corrected hyperspectral spectrum of the grafting point area. Finally, based on the corrected hyperspectral image of the graft union region, the moisture index of the graft union region was calculated using a specific moisture index calculation formula (e.g., NDWI = (R_860 - R_1240) / (R_860 + R_1240), where R_860 and R_1240 are the reflection intensities at wavelengths of 860 nm and 1240 nm, respectively). This method provides a more accurate and reliable moisture index for subsequent assessment of graft union healing quality.

[0049] Specifically, based on the aforementioned infrared image sequence, the steps for calculating the temperature recovery rate of the grafting interface region include: Each frame of the infrared image sequence is divided into sub-regions to form multiple sub-region infrared image sequences. The temperature recovery rate of each sub-region is calculated based on the infrared image sequences of the multiple sub-regions. Based on the temperature recovery rate of each sub-region and its corresponding area weight, a weighted average temperature recovery rate is calculated and used as the temperature recovery rate of the grafting interface region.

[0050] The sub-region division of each frame in the infrared image sequence refers to dividing the overall image of the graft union region into several smaller, independent sub-regions. This division can be performed in various ways; for example, it can be based on a preset grid for uniform division, or it can be adaptively divided according to the geometry or healing characteristics of the graft union. Through this division, multiple sub-region infrared image sequences can be formed, each sequence corresponding to the physiological response of a sub-region after temperature stimulation.

[0051] Furthermore, based on the infrared image sequences of the multiple sub-regions, the temperature recovery rate of each sub-region is calculated. This means analyzing the temperature changes of each independent sub-region after temperature stimulation and calculating the unique temperature recovery rate of that sub-region. This allows the healing dynamics of each local area to be quantified individually.

[0052] In practical applications, a weighted average temperature recovery rate is calculated based on the temperature recovery rate of each sub-region and its corresponding area weight. This average is then used as the temperature recovery rate of the grafting interface region. This means that the calculated temperature recovery rates of each sub-region are combined with their area proportion or other importance weights within the entire grafting interface region, and then weighted averaged. This weighted average can more comprehensively and accurately reflect the overall temperature recovery characteristics of the entire grafting interface region, while also taking into account the differences in local areas.

[0053] The proposed method meticulously divides the grafting interface region into multiple sub-regions and calculates the temperature recovery rate of each sub-region separately, thereby capturing local differences in the healing process within the grafting interface. This refined analysis avoids the information loss caused by treating the entire grafting interface region as a uniform whole. Furthermore, by weighted averaging the temperature recovery rates of each sub-region, the contribution of different sub-regions to the overall healing quality can be comprehensively considered, making the final calculated temperature recovery rate of the grafting interface region more representative and accurate. For example, poorly healed sub-regions may exhibit abnormal temperature recovery rates; through this meticulous division and weighting, these local anomalies can be more effectively identified and quantified, thus providing more reliable data support for subsequent healing quality assessment.

[0054] In some preferred embodiments, it is assumed that the grafting interface area of ​​a Camellia oleifera grafted seedling is being evaluated. First, an infrared image sequence of the grafting interface area after temperature stimulation is acquired using an infrared thermal imager. Then, each frame of the infrared image sequence is divided into, for example, nine sub-regions of 3x3. For each of these nine sub-regions, its temperature change curve after stimulation is extracted, and its respective temperature recovery rate is calculated. For example, if the healing condition of a certain sub-region is poor, its temperature recovery rate may be significantly lower than that of other healthy sub-regions. Finally, based on the area of ​​these nine sub-regions, a weighted average temperature recovery rate is calculated as the final temperature recovery rate of the entire grafting interface area. This method can effectively identify potential localized poor healing within the grafting interface area, thus providing more accurate evaluation results.

[0055] In some embodiments, the step of calculating the temperature recovery rate of each sub-region in the above-mentioned plurality of sub-region infrared image sequences includes: Obtain the ambient temperature when a preset temperature stimulus is applied to the grafting interface area; Extract the curve of the average surface temperature of the grafting interface region over time from the infrared image sequence of the sub-region; Based on the curve, the temperature recovery rate of the sub-region is calculated by fitting an exponential recovery model, and the temperature recovery rate of the sub-region is calculated according to the following formula: T(t) = T_env - (T_env - T_min) exp(-kt) Where T(t) is the surface temperature at time t, T_env is the ambient temperature, T_min is the lowest temperature that the grafting interface region can reach after applying a preset temperature stimulus, k is the temperature recovery rate, and t is time.

[0056] Specifically, after applying a preset temperature stimulus to the grafting interface area, it is necessary to obtain the ambient temperature T_env at the time of the stimulus. This ambient temperature T_env can be obtained in real time by an ambient temperature sensor placed near the grafted seedling, or determined by preset constant environmental conditions. Further, from the infrared image sequence of the sub-region, the curve of the average surface temperature of each sub-region changing over time can be extracted. This is typically achieved by performing pixel-level temperature extraction on each frame of the infrared image sequence and averaging the pixel temperatures within each sub-region.

[0057] The exponential recovery model T(t) = T_env - (T_env - T_min)exp(-kt), i.e., k = 1 / tln((Tenv - T(t) / (Tenv - Tmin)), describes how the temperature gradually recovers to the ambient temperature T_env over time t. In this model, T(t) represents the surface temperature at time t, and T_min represents the lowest temperature that the grafting interface region can reach after applying a preset temperature stimulus, reflecting the intensity of the stimulus and the initial response of the grafting interface. The parameter k is the temperature recovery rate, which physically characterizes how fast the temperature recovers; the larger the value of k, the faster the temperature recovery. By fitting the curve of the actual measured average surface temperature changing over time to this exponential recovery model, the parameter k, i.e., the temperature recovery rate of the sub-region, can be accurately calculated. The fitting process can be implemented using the least squares method or other optimization algorithms to find the optimal value of k.

[0058] This application's solution, by introducing an exponential recovery model, can more accurately capture the physiological response characteristics of the grafting interface region after temperature stimulation. When the grafting interface region is stimulated by temperature, its surface temperature changes and gradually recovers to the ambient temperature over time. This recovery process is not a simple linear change, but usually follows an exponential decay law, which is consistent with the thermophysiological characteristics of organisms. By obtaining the ambient temperature T_env and the lowest temperature T_min reached by the grafting interface region, and combining it with the curve of the actual measured surface temperature change over time, the exponential recovery model can be used for fitting, which can effectively filter out measurement noise and accurately quantify the rate of temperature recovery k from a physical mechanism perspective. The accurate calculation of parameter k provides a more solid physiological basis and higher reliability for assessing the quality of grafting interface healing.

[0059] In some preferred embodiments, it is assumed that a low-temperature stimulus is applied to the grafting interface area of ​​the Camellia oleifera grafted seedling, causing its surface temperature to decrease. After the stimulus ends, an infrared image sequence of the grafting interface area is continuously acquired using an infrared thermal imager. Simultaneously, the ambient temperature T_env is recorded as 25℃. The curve of the average surface temperature of a sub-region changing over time is extracted from the infrared image sequence. For example, after the stimulus ends, the lowest temperature T_min in this sub-region reaches 10℃. Subsequently, the surface temperature of this sub-region gradually recovers, for example, to 15℃ at t=10 seconds and 20℃ at t=20 seconds. Substituting these data points into the exponential recovery model T(t) = T_env - (T_env - T_min) exp(-kt), i.e., T(t) = 25 - (25 - 10) exp(-kt), the temperature recovery rate k of this sub-region can be accurately calculated using a nonlinear fitting algorithm. For example, the fitting may yield a k value of 0.05 s. -1 This k value serves as the temperature recovery rate for that sub-region and is used for subsequent assessment of the graft union healing quality.

[0060] Specifically, in the above-mentioned method for evaluating the quality of Camellia oleifera grafting, the step of evaluating the uniformity of temperature distribution in the grafting area can be further refined.

[0061] The steps for evaluating the uniformity of temperature distribution based on the infrared image sequence include: Based on the infrared image sequence, obtain multiple surface temperatures of the multiple sub-regions after a period of time following stimulation; Calculate the standard deviation of the surface temperature based on the multiple surface temperatures; The surface temperature standard deviation is used as an index of temperature uniformity to evaluate the uniformity of temperature distribution.

[0062] The infrared image sequence refers to a continuous sequence of infrared images of the physiological response obtained after applying a preset temperature stimulus to the grafting interface region, recording the surface temperature distribution of the grafting interface region at different time points. The multiple sub-regions refer to dividing the grafting interface region into several smaller, independent regions for local temperature analysis. By dividing these sub-regions, the temperature changes and distribution characteristics within the grafting interface region can be captured more precisely. The multiple surface temperatures over a period of time after stimulation refer to the surface temperature data of each sub-region extracted from the infrared image sequence within a preset time period after the application of temperature stimulation. These data reflect the physiological response of each sub-region after temperature stimulation.

[0063] Furthermore, the surface temperature standard deviation refers to the statistical calculation of multiple surface temperatures acquired over a period of time after stimulation in the multiple sub-regions, quantifying the dispersion of these temperature data relative to their average value. A smaller standard deviation indicates that the surface temperatures of each sub-region are closer, i.e., the temperature distribution is more uniform; conversely, a larger standard deviation indicates a more uneven temperature distribution. Therefore, using the surface temperature standard deviation as a temperature uniformity index can objectively reflect the temperature distribution in the grafting interface area, thereby assessing its uniformity.

[0064] This application's method quantifies the uniformity of the physiological response of the grafting interface region after temperature stimulation by dividing the region into sub-regions and calculating the standard deviation of the surface temperature of each sub-region. The healing quality of the grafting interface is closely related to the integrity of the tissue structure and the coordination of physiological functions. A well-healed grafting interface typically has a more complete and uniform internal tissue structure and vascular connection. Therefore, when stimulated by external temperature, its heat conduction and heat dissipation capabilities tend to be more consistent, exhibiting a more uniform temperature distribution and recovery pattern. By calculating the standard deviation of surface temperature, this spatial uniformity difference can be accurately captured, thus providing an objective and quantitative assessment basis for the healing quality of the grafting interface.

[0065] The above technical solution provides a more refined and quantitative method for assessing the uniformity of temperature distribution at the grafting interface. Using the standard deviation of surface temperature as an indicator of temperature uniformity effectively reflects the spatial consistency of the internal tissue structure and physiological functions within the grafting interface area. This quantitative assessment method helps avoid errors from subjective judgment, improves the accuracy and reliability of grafting quality assessment, and thus provides more scientific guidance for the cultivation and management of grafted Camellia oleifera seedlings.

[0066] In some embodiments described above, the moisture index of the grafting interface region is obtained by extracting the reflectance intensity at a specific wavelength and calculating it. However, in practical applications, the actual moisture status and hyperspectral response of the plant grafting interface region may be affected by ambient humidity. For example, when the ambient humidity is high, the plant's transpiration will be reduced, which may result in a higher moisture index on the surface of the grafting interface region, but this does not necessarily indicate good healing inside the grafting interface; it may instead mask potential water transport barriers. If the above problems are not addressed, the moisture index calculated solely from spectral data may not accurately reflect the true healing quality of the grafting interface, thus affecting the reliability of the assessment. Therefore, this application further proposes a scheme to correct the moisture index of the grafting interface region for ambient humidity to improve the accuracy of the moisture index.

[0067] Following the step of calculating the moisture index of the grafting interface region based on the reflection intensity R_1300 and the reflection intensity R_1450 according to the above method, the method further includes: Obtain ambient humidity; The moisture index of the grafting interface area is adjusted according to the ambient humidity. When the ambient humidity is greater than a preset humidity threshold, the moisture index is lowered; when the ambient humidity is less than the preset humidity threshold, the moisture index is raised.

[0068] Specifically, obtaining ambient humidity refers to measuring the relative humidity of the environment surrounding the grafted seedling in real time or periodically using devices such as humidity sensors. This ambient humidity data is used for subsequent correction of the moisture index. Correcting the moisture index of the grafting area based on ambient humidity aims to eliminate or reduce the interference of ambient humidity on the spectral moisture index measurement, making the obtained moisture index more accurately reflect the true moisture status of the grafting area. Specifically, when the ambient humidity is greater than a preset humidity threshold, it indicates a high moisture content in the environment. In this case, plant transpiration may be inhibited, resulting in a relatively high moisture index on the surface of the grafting area. To avoid false overestimation caused by this environmental factor, the calculated moisture index needs to be lowered. Conversely, when the ambient humidity is less than the preset humidity threshold, it indicates a drier environment. Plant transpiration may be enhanced, leading to faster surface moisture loss from the grafting area, resulting in a relatively low moisture index on the spectrum. To avoid false underestimation caused by this environmental factor, the calculated moisture index needs to be higher. In practical applications, the preset humidity threshold can be set based on the physiological characteristics, growth stage, and experimental data of the grafted camellia seedlings. For example, it can be set to a relative humidity of 70% or 80%. Adjusting the humidity level can be done by multiplying by a correction factor, subtracting or adding a correction amount, or by using a pre-established correction model. For instance, a regression model based on ambient humidity and actual moisture conditions can be established to achieve more accurate correction.

[0069] This application's solution corrects the moisture index of the grafting interface region by incorporating environmental humidity, effectively compensating for the influence of environmental factors on spectral measurement results. The moisture index of the grafting interface region is a key indicator for assessing its healing quality, and the plant's moisture status is closely related to environmental humidity. When environmental humidity is high, plant transpiration weakens, and water is not easily lost from the leaf or grafting interface surface, which may lead to the hyperspectral spectrum reflecting a higher moisture content than the actual water transport capacity inside the grafting interface. By lowering the moisture index, this overestimation caused by high humidity can be corrected, making it closer to the true healing state of the grafting interface. Conversely, when environmental humidity is low, plant transpiration increases, and water loss accelerates, which may lead to the hyperspectral spectrum reflecting a lower moisture content than the actual moisture status inside the grafting interface. By raising the moisture index, this underestimation caused by low humidity can be compensated for. It is precisely this dynamic correction based on environmental humidity that allows the acquired moisture index to more accurately reflect the physiological moisture status of the grafting interface region, thus providing a more reliable data basis for subsequent healing quality assessment.

[0070] In some embodiments described above, the healing quality of the graft union is judged based on the moisture index, temperature recovery rate, or temperature distribution uniformity of the graft union area. However, in practical applications, the physiological indicators of grafted Camellia oleifera seedlings are affected by various factors, such as the Camellia oleifera variety, growth stage, and environmental humidity. If these factors are not considered, and judgments are made directly based on the absolute values ​​of a single or few indicators, it may lead to biased evaluation results, affecting the accuracy and reliability of grafting quality assessment.

[0071] In this regard, this application further proposes the following steps for judging the healing quality of the grafting interface: Obtain information on the variety, growth stage, and environmental humidity of the grafted seedlings; Based on the variety information, the growth stage information, and the environmental humidity information, the moisture index of the healthy grafting interface, the temperature recovery rate of the healthy grafting interface, and the temperature distribution uniformity of the healthy grafting interface are determined. The difference between the moisture index and the moisture index of the healthy graft interface is calculated to obtain a first difference value; the difference between the temperature recovery rate and the temperature recovery rate of the healthy graft interface is calculated to obtain a second difference value; and the difference between the temperature distribution uniformity and the temperature distribution uniformity of the healthy graft interface is calculated to obtain a third difference value. The healing quality of the grafting interface is determined based on the first difference, the second difference, and the third difference.

[0072] Specifically, obtaining information on the variety, growth stage, and environmental humidity of grafted seedlings refers to acquiring detailed background data on the seedlings to be evaluated through methods such as manual input, sensor data collection, or database queries. Variety information can include the specific cultivated variety of Camellia oleifera, growth stage information refers to the specific time point or developmental state of the seedling from grafting to the healing process, and environmental humidity information can be obtained in real time through environmental sensors.

[0073] Furthermore, based on the obtained variety information, growth stage information, and environmental humidity information, the corresponding healthy graft union moisture index, healthy graft union temperature recovery rate, and healthy graft union temperature distribution uniformity can be determined. These health reference values ​​can be obtained in advance through a database or model established from a large amount of experimental data. This database or model can provide corresponding ranges or benchmark values ​​of health physiological indicators based on different varieties, growth stages, and environmental humidity conditions.

[0074] Subsequently, the measured moisture index of the graft union area was compared with the corresponding moisture index of a healthy graft union to calculate the first difference. Similarly, the measured temperature recovery rate of the graft union area was compared with the corresponding temperature recovery rate of a healthy graft union to calculate the second difference. Simultaneously, the assessed temperature distribution uniformity was compared with the corresponding temperature distribution uniformity of a healthy graft union to calculate the third difference. These differences reflect the degree of deviation between the physiological state and the healthy state of the graft union being evaluated.

[0075] Finally, the healing quality of the graft union is comprehensively judged based on the calculated first, second, and third differences. For example, a threshold can be set so that if any difference exceeds a preset range, it can be determined that there is a problem with the graft union healing.

[0076] This application's solution addresses the inaccuracy of traditional methods in complex and variable environments by incorporating information on the grafted seedling's variety, growth stage, and environmental humidity to determine reference indicators for a healthy graft union. Specifically, the physiological state of a grafted seedling is influenced by its genetic characteristics (variety), developmental process (growth stage), and external environment (humidity). By acquiring this contextual information, a personalized "health baseline" can be established for the graft union to be evaluated. When the actual measured moisture index, temperature recovery rate, and temperature distribution uniformity are compared to this baseline, the calculated differences more accurately reflect whether abnormal healing has occurred at the graft union, rather than simply fluctuations in indicators caused by normal physiological variations or environmental factors. This difference-based judgment mechanism makes the evaluation results more robust and targeted.

[0077] Through the above technical solution, this application can significantly improve the accuracy and reliability of Camellia oleifera grafting quality assessment. By considering multiple key factors affecting the physiological state of grafted seedlings, the determined health reference indicators are more targeted, avoiding misjudgments caused by varietal differences, changes in growth stage, or environmental fluctuations. Therefore, poorly healed graft unions can be identified more accurately, providing a scientific basis for subsequent agricultural management and intervention measures, effectively reducing false positive or false negative assessment results, thereby improving the survival rate and production efficiency of grafted seedlings.

[0078] Finally, these differences are compared with preset thresholds. For example, if the preset thresholds for moisture index difference are ±0.05, temperature recovery rate difference are ±0.01, and temperature distribution uniformity difference are ±0.5, in this example, the first difference (-0.07) exceeds the moisture index difference threshold, and the third difference (0.7) exceeds the temperature distribution uniformity difference threshold. Therefore, it can be determined that the healing quality of the grafting joint is problematic, possibly indicating poor healing or pseudo-healing. This multi-dimensional, context-aware differential judgment makes the evaluation results more accurate.

[0079] In some embodiments described above, the healing quality of the graft union can be assessed by calculating the differences between the moisture index, temperature recovery rate, and temperature distribution uniformity of the graft union area and their corresponding indicators from those of a healthy graft union. However, in practical applications, these differences alone may not be sufficient to accurately distinguish between different types of healing problems, such as pseudo-healing, thus affecting the accuracy and guidance of the assessment. Therefore, this application further proposes setting specific threshold conditions to identify pseudo-healing when assessing the healing quality of the graft union.

[0080] In this regard, this application further proposes steps for judging the healing quality of the grafting interface, including: When the first difference is greater than the preset moisture index difference threshold, or the second difference is greater than the temperature recovery rate difference threshold, or the third difference is less than the temperature distribution uniformity threshold, the grafting interface is determined to be a false healing.

[0081] Specifically, the "preset moisture index difference threshold," "temperature recovery rate difference threshold," and "temperature distribution uniformity threshold" are critical values ​​pre-set based on factors such as the variety information, growth stage information, and environmental humidity information of the grafted Camellia oleifera seedlings, through extensive experimental data, historical records, or expert experience. These thresholds are used to define the degree to which various physiological indicators of the graft union deviate from a healthy state, thereby determining whether there is poor healing. Among them, "false healing" can be understood as the graft union showing signs of healing on the outside, but its internal tissue connection is insufficient or functional healing is incomplete, which affects the long-term growth of the grafted seedling and may even lead to eventual failure.

[0082] This application's solution, by introducing specific threshold conditions, enables a more refined assessment of the healing quality of the graft union, particularly in identifying false healing. When the difference between the moisture index of the graft union and that of a healthy graft union (the first difference) exceeds a preset moisture index difference threshold, it indicates a potential serious impairment in the moisture transport capacity of the graft union area, a key physiological manifestation of false healing. Similarly, when the difference between the temperature recovery rate of the graft union and that of a healthy graft union (the second difference) exceeds a preset temperature recovery rate difference threshold, it suggests abnormal physiological metabolic activity in the graft union area, with heat loss or impaired recovery mechanisms—also typical characteristics of false healing. Furthermore, when the temperature distribution uniformity of the graft union area (the third difference) is lower than a preset temperature distribution uniformity threshold, it indicates uneven tissue connection or healing process within the graft union, potentially leading to poor local healing or cavities, thus resulting in false healing. By comprehensively assessing the deviations of these physiological indicators, this application can effectively reveal the true healing status of the graft union from multiple dimensions, avoiding misjudgments that may arise from relying solely on appearance.

[0083] In large-scale camellia oleifera cultivation, to improve efficiency and standardization, traditional grafting quality assessment systems have evolved. Computer vision-based systems were originally designed to quickly distinguish between "obvious survival" and "obvious failure" at the graft union. However, with the diversification of grafting material sources, a phenomenon called "false healing" has become increasingly common. In these cases, the graft union appears healthy on the outside, but the internal channels responsible for transporting water and nutrients are not fully connected or are not firmly connected. Existing assessment systems, relying solely on surface appearances, struggle to detect these internally problematic seedlings, leading to them being mistakenly treated as healthy and ultimately resulting in unexpectedly low survival rates during subsequent growth.

[0084] In response, this application proposes a Camellia oleifera grafting quality assessment system for evaluating the graft union healing quality of Camellia oleifera grafted seedlings. The system includes: The hyperspectral image acquisition module is used to collect the hyperspectral image of the grafting interface region in a preset band and obtain the hyperspectral image of the grafting interface region. The moisture index calculation module is used to obtain the moisture index of the grafting interface region based on the hyperspectral spectrum of the grafting interface region. An infrared image sequence acquisition module is used to apply a preset temperature stimulus to the grafting interface region and acquire an infrared image sequence of the physiological response of the grafting interface region after the temperature stimulus. The temperature recovery rate calculation module is used to calculate the temperature recovery rate of the grafting interface region based on the infrared image sequence. A temperature distribution uniformity assessment module is used to assess the temperature distribution uniformity based on the infrared image sequence. The judgment module is used to judge the healing quality of the grafting interface based on the moisture index, the temperature recovery rate, or the temperature distribution uniformity.

[0085] The Camellia oleifera grafting quality assessment system proposed in this application integrates a hyperspectral image acquisition module, a moisture index calculation module, an infrared image sequence acquisition module, a temperature recovery rate calculation module, a temperature distribution uniformity assessment module, and a final judgment module, constructing a comprehensive, non-contact assessment framework. This system can conduct in-depth analysis of the graft union healing quality of Camellia oleifera grafted seedlings from multiple physiological dimensions, effectively overcoming the limitations of existing technologies that rely solely on surface morphology observation. It exhibits a significant advantage, particularly in identifying "false healing" phenomena, thus providing reliable technical support for improving the survival rate and planting benefits of Camellia oleifera grafted seedlings.

[0086] The moisture index calculation module is designed to obtain the moisture index of the grafting interface region based on its hyperspectral image. This module can consist of a data processor (e.g., a microcontroller, FPGA, or application-specific integrated circuit) and an algorithm program stored within it. The data processor receives the hyperspectral image data and executes a preset algorithm to calculate the moisture index. For example, a preset algorithm can extract the reflection intensity of specific wavelengths (such as 1300 nm and 1450 nm) from the hyperspectral image and calculate the moisture index based on these intensities.

[0087] The infrared image sequence acquisition module applies a preset temperature stimulus to the graft union area and acquires an infrared image sequence of the physiological response of the graft union area after the temperature stimulus. This module may include a temperature stimulation device and an infrared thermal imager. The temperature stimulation device may be a small heater or cooling device used to locally change the temperature of the graft union area within a short period of time. The infrared thermal imager is configured to continuously capture infrared images of the graft union area at a certain frame rate (e.g., 10 frames per second) to record the sequence of its surface temperature changes over time.

[0088] The temperature recovery rate calculation module calculates the temperature recovery rate of the grafting interface region based on an infrared image sequence. This module can consist of a data processor and a temperature recovery rate calculation algorithm stored within it. The data processor receives the infrared image sequence and extracts a curve showing the change in the average surface temperature of the grafting interface region over time. Subsequently, it calculates the temperature recovery rate of the grafting interface region by executing a mathematical fitting algorithm (e.g., an exponential recovery model).

[0089] The temperature distribution uniformity assessment module evaluates the temperature distribution uniformity based on an infrared image sequence. This module can consist of a data processor and a uniformity assessment algorithm stored within it. The data processor selects multiple frames of images from the infrared image sequence over a period of time following the stimulus and calculates the surface temperature of different sub-regions within the grafting interface area. Then, by executing a statistical analysis algorithm, such as calculating the standard deviation of the surface temperatures of these sub-regions, and using this standard deviation as a temperature uniformity index, the temperature distribution uniformity of the grafting interface area is assessed.

[0090] The judgment module determines the healing quality of the graft union based on moisture index, temperature recovery rate, or temperature distribution uniformity. This module can consist of a data processor and a decision algorithm stored within it. The data processor receives the three indicators—moisture index, temperature recovery rate, and temperature distribution uniformity—and compares them with preset healthy graft union standard values. Based on these comparisons, the decision algorithm outputs a healing quality judgment result for the graft union, such as "healthy healing," "false healing," or "healing failure."

[0091] The Camellia oleifera grafting quality assessment system proposed in this application aims to address the shortcomings of traditional assessment systems in identifying the phenomenon of "false healing" in grafted Camellia oleifera seedlings. Existing computer vision-based assessment systems primarily focus on identifying the external morphological characteristics of the graft union, such as color and degree of swelling. However, these surface features often fail to accurately reflect the connection status and physiological function of the vascular bundles within the graft union. When false healing occurs, the external appearance of the graft union appears normal, but the internal transport of water and nutrients is obstructed, leading to a decrease in the survival rate later on.

[0092] To address this, the evaluation system of this application introduces a hyperspectral mapping module and an infrared image sequence acquisition module, enabling analysis of the grafting interface from a deeper physiological perspective. The hyperspectral mapping module provides detailed spectral information of the grafting interface region, quantifying its water content through a moisture index calculation module, thereby reflecting the internal water supply status. The infrared image sequence acquisition module, combined with temperature stimulation, dynamically captures heat conduction and metabolic activity in the grafting interface region through a temperature recovery rate calculation module and a temperature distribution uniformity assessment module, revealing the patency and uniformity of vascular bundle connections.

[0093] Compared with existing technologies, the system in this application no longer relies solely on surface morphology, but comprehensively utilizes multi-dimensional physiological indicators provided by hyperspectral and infrared thermal imaging technologies. This integrated evaluation method enables the system to effectively identify the "false healing" phenomenon that is difficult to detect using traditional methods, significantly improving the accuracy and reliability of the evaluation. Through more precise quality assessment, seedlings with false healing can be avoided from being misjudged as healthy seedlings, thereby improving the overall survival rate of grafted Camellia oleifera seedlings, reducing planting risks, and bringing substantial technological progress and economic benefits to the large-scale cultivation of Camellia oleifera.

[0094] The foregoing has provided a detailed description of the preferred embodiments of this application. However, this application is not limited to the above-described embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application. All such equivalent modifications or substitutions are included within the scope defined in this application.

Claims

1. A method for evaluating the grafting quality of a Camellia oleifera grafting seedling, for evaluating the healing quality of a grafting interface of a Camellia oleifera grafting seedling, characterized in that, The method comprises the following steps: Collecting a hyperspectral spectrum of the grafting interface region in a preset wave band to obtain a hyperspectral spectrum of the grafting interface region; Obtaining a water index of the grafting interface region according to the hyperspectral spectrum of the grafting interface region; Applying a preset temperature stimulus to the grafting interface region to obtain a physiological response infrared image sequence of the grafting interface region after the temperature stimulus; Calculating a temperature recovery rate of the grafting interface region according to the infrared image sequence; Evaluating temperature distribution uniformity according to the infrared image sequence; Judging the healing quality of the grafting interface according to the water index, the temperature recovery rate or the temperature distribution uniformity.

2. The method of claim 1, wherein, The step of obtaining the water index of the grafting interface region according to the hyperspectral spectrum of the grafting interface region comprises: Extracting a reflection intensity R_1300 at a wavelength of 1300 nm and a reflection intensity R_1450 at a wavelength of 1450 nm from the hyperspectral spectrum of the grafting interface region; Calculating the water index of the grafting interface region according to the reflection intensity R_1300 and the reflection intensity R_1450, wherein the water index of the grafting interface region = (R_1300 - R_1450) / (R_1300 + R_1450).

3. The method of claim 1, wherein, The step of obtaining the water index of the grafting interface region according to the hyperspectral spectrum of the grafting interface region comprises: Obtaining a hyperspectral spectrum of a healthy stem section within a certain range of the grafting interface region, wherein the certain periphery is a region 10 cm above and below the grafting interface region; Correcting the hyperspectral spectrum of the grafting interface region according to the hyperspectral spectrum of the healthy stem section to obtain a corrected hyperspectral spectrum of the grafting interface region; Obtaining the water index of the grafting interface region according to the corrected hyperspectral spectrum of the grafting interface region.

4. The method of claim 1, wherein, The step of calculating the temperature recovery rate of the grafting interface region according to the infrared image sequence comprises: Dividing each frame of image in the infrared image sequence into sub-regions to form a plurality of sub-regional infrared image sequences; Calculating the temperature recovery rate of each sub-region according to the plurality of sub-regional infrared image sequences; Calculating a weighted average temperature recovery rate as the temperature recovery rate of the grafting interface region according to the temperature recovery rate of each sub-region and the corresponding area weight.

5. The method of claim 4, wherein, The step of calculating the temperature recovery rate of each sub-region according to the plurality of sub-regional infrared image sequences comprises: Obtaining an environmental temperature when the preset temperature stimulus is applied to the grafting interface region; Extracting a curve of the average surface temperature of the grafting interface region changing with time from the sub-regional infrared image sequence; Calculating the temperature recovery rate of the sub-region by fitting an exponential recovery model according to the curve, wherein the temperature recovery rate of the sub-region is calculated according to the following formula: T(t) = T_env - (T_env - T_min) exp(-kt) wherein T(t) is the surface temperature at time t, T_env is the environmental temperature, T_min is the lowest temperature that the grafting interface region can reach after the preset temperature stimulus is applied, k is the temperature recovery rate, and t is the time.

6. The method of claim 1, wherein, The step of evaluating the uniformity of the temperature distribution according to the infrared image sequence comprises: According to the infrared image sequence, a plurality of surface temperatures of a plurality of sub-regions at a time after stimulation are obtained; According to the plurality of surface temperatures, a surface temperature standard deviation is calculated; The surface temperature standard deviation is taken as a temperature uniformity index to evaluate the uniformity of the temperature distribution.

7. The method of claim 2, wherein, The step of calculating the moisture index of the grafting interface region according to the reflection intensity R_1300 and the reflection intensity R_1450 further comprises: Obtaining the environmental humidity; According to the environmental humidity, the moisture index of the grafting interface region is corrected, when the environmental humidity is greater than a preset humidity threshold, the moisture index is lowered, and when the environmental humidity is less than a preset humidity threshold, the moisture index is increased.

8. The method of claim 1, wherein, The step of judging the healing quality of the grafting interface according to the moisture index, the temperature recovery rate, or the temperature distribution uniformity comprises: Obtaining the variety information, growth stage information, and environmental humidity information of the grafted seedling; According to the variety information, the growth stage information, and the environmental humidity information, the healthy grafting interface moisture index, the healthy grafting interface temperature recovery rate, and the healthy grafting interface temperature distribution uniformity are determined; The difference between the moisture index and the healthy grafting interface moisture index is calculated to obtain a first difference value, the difference between the temperature recovery rate and the healthy grafting interface temperature recovery rate is calculated to obtain a second difference value, and the difference between the temperature distribution uniformity and the healthy grafting interface temperature distribution uniformity is calculated to obtain a third difference value; According to the first difference value, the second difference value, and the third difference value, the healing quality of the grafting interface is judged.

9. The method of claim 8, wherein, The step of judging the healing quality of the grafting interface according to the first difference value, the second difference value, and the third difference value comprises: When the first difference value is greater than a preset moisture index difference threshold or the second difference value is greater than a temperature recovery rate difference threshold or the third difference value is less than a temperature distribution uniformity threshold, the grafting interface is determined to be a false healing.

10. A Camellia oleifera grafting quality evaluation system for evaluating the grafting interface healing quality of Camellia oleifera grafted seedlings, characterized in that, The system comprises: A hyperspectral spectrum acquisition module is configured to collect a hyperspectral spectrum of a grafting interface region at a preset waveband to obtain a hyperspectral spectrum of the grafting interface region; A moisture index calculation module is configured to obtain a moisture index of the grafting interface region according to the hyperspectral spectrum of the grafting interface region; An infrared image sequence acquisition module is configured to apply a preset temperature stimulus to the grafting interface region and obtain a physiological response infrared image sequence of the grafting interface region after the temperature stimulus; A temperature recovery rate calculation module is configured to calculate a temperature recovery rate of the grafting interface region according to the infrared image sequence; A temperature distribution uniformity evaluation module is configured to evaluate the uniformity of the temperature distribution according to the infrared image sequence; A judgment module is configured to judge the healing quality of the grafting interface according to the moisture index, the temperature recovery rate, or the temperature distribution uniformity.

Images