Video melting point acquisition method, melting point instrument, program product and storage medium

Abstract

The invention discloses a video melting point acquisition method, a melting point instrument, a program product and a storage medium, and relates to the field of analyzing materials by measuring physical properties of the materials. The method comprises the following steps: identifying sublimation pixel points generated in sample temperature rise through image difference; when the number exceeds a threshold value, starting linear heating, and continuously analyzing the image to generate a hyperbolic curve of the main light transmittance and the main texture complexity; if the light transmittance curve has a negative slope and the texture complexity curve does not decline, excluding the section of abnormal data; determining the temperature at the moment when the slope of the light transmittance curve exceeds a forward threshold for the first time and the texture complexity curve synchronously starts to drop as the initial melting point of the sample; and determining the temperature when the slope of the light transmittance curve tends to be gentle and the texture complexity curve is stabilized at the minimum value as the final melting point of the sample. By implementing the technical scheme provided by the invention, the accuracy of an automatic melting point measurement result is improved.

Description

Technical Field

[0001] This application relates to the field of materials analysis by measuring the physical properties of materials, and more particularly to a video melting point acquisition method, a melting point apparatus, a program product, and a storage medium. Background Technology

[0002] Currently, melting point, as one of the fundamental physical properties of matter, is a core indicator characterizing the purity and identity of a substance. With the rapid development of pharmaceutical, chemical, and new materials science fields, measuring melting point has become crucial, and automated melting point measurement technology is widely used in modern industrial and scientific research scenarios.

[0003] In related technologies, automated melting point measurement typically employs dynamic analysis methods based on video image processing. First, the initial contour of the sample to be tested is identified and defined as the Region of Interest (ROI). During programmed heating, edge detection or target tracking algorithms are used to update the boundaries of this ROI in real time to accommodate potential volume shrinkage of the sample during heating, ensuring the measurement remains focused on the sample itself. Finally, the start and end of the melting process are determined by continuously calculating the change curves of the average transmittance or pixel grayscale values ​​within this dynamic region.

[0004] However, when certain crystalline or powder samples undergo a dramatic solid-state transformation before melting, such as the loss of internal water of crystallization or crystal rearrangement, the outer contour of the sample may not change significantly, but new shadows or tiny dark areas may form inside due to compaction or collapse. In this case, although the dynamically adjusted measurement area can accurately frame the sample, the overall average transmittance inside will show an unrealistic decrease. This signal disturbance caused by changes in internal physical morphology rather than a true phase transition is difficult to distinguish from genuine pre-melting phenomena, which may affect the accuracy of the initial melting point determination results. Summary of the Invention

[0005] This application provides a video melting point acquisition method, a melting point apparatus, a program product, and a storage medium to improve the accuracy of automatic melting point determination results.

[0006] The first aspect of this application provides a video melting point acquisition method, the method comprising: The sample area is heated from its initial temperature to a first preset temperature while continuously acquiring real-time images. By performing a difference operation between the real-time images and a reference image, newly generated pixels above the sample area are identified. When the number of new pixels exceeds a preset pixel threshold, image data analysis is performed on the sample area during linear heating to generate a main transmittance curve and a main texture complexity curve. When a negative slope interval is identified in the main transmittance curve, and the main texture complexity curve shows a non-decreasing trend within the temperature range corresponding to the negative slope interval, the transmittance curve data corresponding to the negative slope interval is excluded from subsequent calculations. The temperature at which the slope of the main transmittance curve first exceeds a preset first positive threshold and the main texture complexity curve simultaneously begins to decrease is determined as the initial melting point of the sample. The temperature at which the slope of the main transmittance curve approaches zero and the main texture complexity curve stabilizes at its lowest value plateau is determined as the final melting point of the sample.

[0007] In the above embodiments, the principal texture complexity curve and the principal transmittance curve are introduced for collaborative analysis. When it is identified that the transmittance decreases in a non-true manner due to solid-phase transformations such as sample collapse (i.e., the principal transmittance curve has a negative slope range), the key feature that the principal texture complexity curve shows a non-decreasing trend at this time is simultaneously judged. This identifies and eliminates artifact data, ensuring that the subsequent determination of the initial melting point is based on the signal changes in the true melting stage, that is, the moment when the transmittance first increases significantly and the texture complexity begins to decrease simultaneously. This solves the technical defect in related technologies that misjudge the initial melting point due to changes in the physical morphology of the sample, and improves the accuracy and reliability of melting point determination.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, the temperature at which the slope of the main transmittance curve first exceeds a preset first positive threshold and the main texture complexity curve simultaneously begins to decrease is determined as the initial melting point of the sample, specifically including: On the main transmittance curve and the main texture complexity curve, all temperature points where the slope of the transmittance curve exceeds the first positive threshold and the texture complexity curve begins to decrease are taken as initial melting candidate points. Starting from each initial melting candidate point, a preset continuous temperature zone is set as a confirmation window to check the steady-fall state that appears within the confirmation window; check the steady-increase state within the confirmation window; and finally determine the first initial melting candidate point that simultaneously satisfies both the steady-fall state and the steady-increase state as the initial melting point of the sample.

[0009] In the above embodiments, the candidate melting points are screened by the synergistic change of transmittance and texture complexity, forming the first layer of screening. Then, a confirmation window mechanism is introduced, which performs a dual test on the persistence and stability of the melting trend within the window. That is, it tests whether the steady decrease of texture complexity and the steady increase of transmittance are satisfied at the same time. This can effectively distinguish the transient and unstable signal fluctuations caused by sample softening, cracking, etc. from the real and irreversible melting process, thereby improving the anti-interference ability and accuracy of melting point detection.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, the first candidate melting point that simultaneously satisfies both the steady-fall state and the steady-rise state is ultimately determined as the initial melting point of the sample, specifically including: The first candidate melting point that simultaneously satisfies both the steady-fall and steady-rise states is identified as the undetermined initial melting point. If recrystallization characteristics are detected within the verification window, the undetermined initial melting point is excluded, and the next candidate melting point is searched from the temperature point after the undetermined initial melting point as the undetermined initial melting point. If no recrystallization characteristics are detected within the verification window, the undetermined initial melting point is finally identified as the initial melting point of the sample.

[0011] In the above embodiments, after the initial melting point to be determined is initially determined through the confirmation window, a verification window is further set to detect whether there are melting reversal characteristics caused by recrystallization in the subsequent temperature range. This can separate the false melting point signal generated by complex phase transformation from the real melting process. By actively eliminating the false melting point, it is ensured that the final determined initial melting point is the irreversible final melting starting point of the sample, thereby improving the accuracy and reliability of melting point determination for samples with complex thermal behavior.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, when the number of new pixels exceeds a preset pixel threshold, before performing image data analysis on the sample region and generating the main transmittance curve and the main texture complexity curve during the linear heating process of the sample region, the method further includes: The moment a new pixel is detected is taken as the starting point for dynamic discrimination. Within a preset discrimination period after the starting point, image data analysis is continuously performed on the area above the new pixel to generate a temporal sequence of texture in the area above. The temporal variance of the temporal sequence of texture in the area above is calculated. When the temporal variance exceeds a preset dynamic threshold, the current linear heating process is stopped and the temperature of the sample area is cooled back to a first preset temperature. At the first preset temperature, continuous acquisition of real-time images of the sample area is restarted. By performing inter-frame difference operations on the real-time images of the re-acquired images, the newly generated re-acquired pixels above the sample area are identified and replaced with the new pixels.

[0013] In the above embodiments, by calculating the temporal variance of the texture time series of the upper region, the dynamic stability of pre-melting phenomena such as sublimation and decomposition can be quantitatively identified. When a violent, unsteady process caused by sudden boiling or bursting is detected (i.e., the temporal variance exceeds the threshold), the heating can be actively interrupted and a cooling and resampling operation can be performed. This filters out the interference of instantaneous physical changes on subsequent melting point determination, ensuring the stability and reliability of the analytical benchmark, thereby improving the universality and accuracy of measurements on samples with complex thermal behavior.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, at a first preset temperature, continuous acquisition of resampled real-time images of the sample area is restarted, and by performing inter-frame difference operations on the resampled real-time images, newly generated resampled pixels above the sample area are identified, and the resampled pixels are used to replace the new pixels, specifically including: The image above the sample area is continuously monitored. When the dynamic droplet pixel transforms into a stable pixel and stabilizes at a first preset temperature, an image containing solid artifacts is acquired as an artifact reference frame. The artifact reference frame is compared with the baseline image to identify the difference pixels. An artifact mask is generated based on the position of the difference pixels. At the first preset temperature, the real-time image of the sample area is continuously acquired again. By performing inter-frame difference operations on the real-time image of the reacquired image, the newly generated reacquired pixels above the sample area are identified. Pixels whose spatial positions coincide with the mask marking positions in the reacquired pixels are removed to obtain corrected reacquired pixels. The corrected reacquired pixels are then used to replace the new pixels.

[0015] In the above embodiments, by pre-generating an artifact mask that marks the location of solid artifacts, the newly appearing pixels can be spatially identified and removed during subsequent resampling. This separates the solid artifact interference remaining from the previous measurement from the real melting signal, eliminates the artifact confusion effect, ensures the purity of the resampling data, and thus guarantees the accuracy and reliability of the final melting point determination.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, when a negative slope interval is identified in the main transmittance curve, and the main texture complexity curve exhibits a non-decreasing trend within the temperature range corresponding to the negative slope interval, after excluding the transmittance curve data corresponding to the negative slope interval in subsequent calculations, the method further includes: When the main texture complexity curve is detected to be continuously decreasing and the main transmittance curve is within the corresponding temperature range, but the slope fails to meet the first positive threshold condition, the initial melting point of the sample is determined only based on the temperature corresponding to the starting moment of the continuous decrease in the main texture complexity curve.

[0017] In the above embodiments, by performing synergistic analysis on the two curves of main texture complexity and main transmittance, when a special melting situation is identified in which the sample structure has begun to collapse (texture complexity continues to decrease) but transmittance has not improved significantly, the judgment weight can be completely assigned to the more essential structural change characteristics, thereby getting rid of the single dependence on transmittance changes, capturing the true melting point of such atypical samples, and expanding the applicability and accuracy of melting point determination.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, after performing image data analysis on the sample area to generate the main transmittance curve and the main texture complexity curve, the method further includes: During the linear heating process, when the low-activity melting state persists for a preset temperature window threshold or a preset time window threshold, the sample region is divided into a grid containing multiple independent sub-regions; the image data of each sub-region is calculated, and a corresponding sub-transmittance curve and sub-texture complexity curve are generated for each sub-region; the temperature at which the slope of the corresponding sub-transmittance curve of each sub-region first exceeds the preset sub-region positive threshold and the corresponding sub-texture complexity curve simultaneously begins to decrease is determined as the true initial melting point.

[0019] In the above embodiments, after identifying a low-activity melting state with a weak overall melting signal, the analysis granularity is reduced from the overall to the local by activating the gridded analysis mode. This allows the sub-region where melting occurs first to be located, solving the problem of overall signal lag and distortion caused by the non-uniformity of sample melting. This ensures the capture of the true initial melting point and improves the sensitivity and reliability of the measurement.

[0020] In a second aspect, embodiments of this application provide a melting point apparatus, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the melting point apparatus to perform the method described in the first aspect and any possible implementation thereof.

[0021] Thirdly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a melting point apparatus, cause the melting point apparatus to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a melting point apparatus, cause the melting point apparatus to perform the method described in the first aspect and any possible implementation thereof.

[0023] Understandably, the melting point apparatus provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the video melting point acquisition method provided in the embodiments of this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. This application introduces a collaborative analysis of the principal texture complexity curve and the principal transmittance curve. When it identifies a non-true decrease in transmittance due to solid-phase transformations such as sample collapse (i.e., the principal transmittance curve exhibits a negative slope), it simultaneously identifies and eliminates artifact data by judging the key characteristic that the principal texture complexity curve shows a non-decreasing trend. This ensures that the subsequent determination of the initial melting point is based on the signal changes during the true melting stage, i.e., the moment when transmittance first significantly increases and texture complexity simultaneously begins to decrease. This solves the technical defect in related technologies where the initial melting point is misjudged due to changes in the physical morphology of the sample, improving the accuracy and reliability of melting point determination.

[0025] 2. This application pre-generates an artifact mask that marks the location of solid artifacts. In the subsequent resampling process, it can spatially identify and remove newly appearing pixels, thereby separating the solid artifact interference remaining from the previous measurement from the real melting signal, eliminating the artifact confusion effect, ensuring the purity of the resampling data, and thus ensuring the accuracy and reliability of the final melting point determination.

[0026] 3. After initially determining the undetermined initial melting point through the confirmation window, this application further sets up a verification window to detect whether there are melting reversal characteristics caused by recrystallization in the subsequent temperature range. This can separate the false melting point signal generated by complex phase transformation from the real melting process. By actively eliminating the false melting point, it ensures that the finally determined initial melting point is the irreversible final melting starting point of the sample, thereby improving the accuracy and reliability of melting point determination for samples with complex thermal behavior. Attached Figure Description

[0027] Figure 1 This is a flowchart illustrating a video melting point acquisition method in an embodiment of this application; Figure 2 This is a schematic diagram of the principal transmittance and principal texture complexity curves during the sample melting process of the video melting point acquisition method in this application embodiment; Figure 3 This is another flowchart illustrating the video melting point acquisition method in this application embodiment; Figure 4 This is a schematic diagram of an exemplary hardware structure of a melting point apparatus in an embodiment of this application. Detailed Implementation

[0028] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0029] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0030] In related technologies, automated melting point measurement mainly relies on image analysis of the sample area, determining the melting point by calculating the overall transmittance change of the sample area during heating. However, before actual melting, some samples undergo solid-state transformations, such as crystal rearrangement or collapse of internal voids. This process causes the sample to become more compacted, forming new shadowed areas, resulting in an abnormal decrease in the overall transmittance detected by the instrument. Because related technologies rely on only a single transmittance index, they cannot effectively distinguish between this spurious signal caused by physical morphological changes and the true pre-melting signs. Therefore, it is very easy to misjudge the starting point of this spurious signal as the initial melting point, leading to significant deviations in the measurement results.

[0031] In this embodiment, to address the aforementioned issues, the solution introduces principal texture complexity as an analytical dimension parallel to principal transmittance. When a negative slope (i.e., a decrease in transmittance) is detected in the principal transmittance curve, the principal texture complexity curve is simultaneously examined. If the texture complexity does not show a decreasing trend at this time, it is determined that the decrease in transmittance is caused by non-melting physical changes such as internal sample collapse, rather than the start of actual melting. Therefore, the data in this negative slope range is excluded, and the process continues until the transmittance first significantly increases and the texture complexity simultaneously begins to decrease, at which point the true initial melting point is determined. This collaborative analysis mechanism can identify and filter out interference from solid-phase transformation, resolving the problem of misjudgment of the initial melting point, thereby improving the accuracy and universality of melting point determination.

[0032] Figure 1 This is a flowchart illustrating the video melting point acquisition method used in the embodiments of this application, including the following steps: S101. Raise the sample area from the initial temperature to the first preset temperature and continuously acquire real-time images. By performing a difference operation between the real-time images and the reference images, identify the newly generated pixels above the sample area.

[0033] The sample region refers to the set of pixels containing the sample powder identified and delineated from the reference image using image segmentation algorithms (such as edge detection); the initial temperature represents a temperature that ensures the sample is in a completely solid and stable state, usually set to room temperature or slightly higher based on common knowledge or safe operating procedures; the first preset temperature refers to a temperature close to the sample's possible sublimation point but below its initial melting point, usually based on empirical judgment of the physical properties of such substances or determined through pre-experiments; the reference image refers to a static image taken at a stable initial temperature; and new pixels represent tiny particles that are not present in the reference image, formed by the recondensation of the sample on a cooler surface after sublimation.

[0034] Specifically, the sample is heated to a first preset temperature and maintained at a constant temperature. During this period, a camera is controlled to continuously capture real-time images of the area above the sample at a high frame rate. By performing pixel-level grayscale difference operations on each real-time image and a reference image in the initial state, minute changes can be highlighted. Physically, as the temperature rises, many organic substances undergo slight sublimation, even far from reaching their melting point. The sublimated gaseous molecules come into contact with the cooler cavity surface above the sample and recondense into solid particles. These particles scatter or block light, thus forming new, slightly bright or slightly dark pixels in the camera image. The difference operation filters out background noise, retaining only these newly added pixels as signals. Therefore, this step utilizes sublimation, a preceding physical phenomenon, as a confirmation signal that the sample has responded to thermal excitation.

[0035] S102. When the number of new pixels exceeds the preset pixel threshold, during the linear heating process of the sample area, image data analysis is performed on the sample area to generate the main transmittance curve and the main texture complexity curve.

[0036] Among them, the preset pixel threshold refers to the critical number of pixels used to distinguish between real sublimation signals and random image noise. It is usually set empirically using, for example, the 3σ principle, through statistical analysis of noise from a large number of blank experiments. Linear heating refers to the heating control system raising the sample stage temperature at a constant rate to establish a linear relationship between temperature and time. The principal transmittance curve represents a quantitative indicator of the overall light transmission capability of the sample area, which is usually obtained by calculating the average gray value of all pixels in the sample area. The principal texture complexity curve represents the degree of change of pixel gray values ​​in the image within the sample area.

[0037] Specifically, when the cumulative number of new pixels detected in S101 exceeds a preset threshold, the sample is determined to have entered a thermally active state, and a linear heating program is initiated. During this process, a two-dimensional synchronous analysis is performed on the sample area of ​​each frame of the image. On the one hand, the principal transmittance is calculated. Physically, this means that there are a large number of scattering interfaces between solid powder particles, making the overall structure opaque or semi-transparent with low transmittance (low average gray value). When it begins to melt into a liquid, the particles disappear, becoming a homogeneous liquid phase, making it easier for light to penetrate, and the transmittance increases significantly (average gray value increases). On the other hand, the principal texture complexity is calculated. Physically, this means that the solid powder has various morphologies, which manifest as highly complex textures with rich details and dramatic gray value changes in the image. After melting, it becomes a smooth liquid surface, and the image texture tends to be simple and homogeneous, with the complexity decreasing sharply. These two curves, from two orthogonal dimensions of optical properties and macroscopic structure, jointly depict the melting process of the sample.

[0038] In some embodiments, the two curves can be generated in multiple ways: Optionally, the grayscale average and grayscale variance are used for calculation: For the main transmittance curve, at each temperature point, the grayscale values ​​of all pixels in the sample area are extracted, and their arithmetic mean is calculated as the transmittance value of that point; For the main texture complexity curve, at each temperature point, the variance or standard deviation of the grayscale values ​​of all pixels in the same sample area is calculated as the texture complexity value of that point; The calculated sequence values ​​are plotted as curves with the corresponding temperature sequences.

[0039] It is understandable that other image statistics or frequency domain analysis methods (such as Fourier transform spectrum analysis) can also be used to characterize transmittance and texture complexity, which is not limited here.

[0040] In some embodiments, after the initial thermal response signal is detected in S101 and before the linear heating in S102, a dynamic discrimination and intelligent reset process can be executed to actively identify and handle boiling or sputtering artifacts caused by residual solvent in the sample, thereby avoiding invalid measurements and realizing process self-repair.

[0041] Specifically, if the sample contains trace amounts of unevaporated solvent, local overheating may occur in the early stages of heating, leading to boiling or splashing, which may spray sample particles or droplets onto the observation window above. This is not a melting phenomenon, and if not properly identified, it will seriously interfere with subsequent transmittance and texture analysis.

[0042] When S101 detects a new pixel for the first time and confirms that the number exceeds the threshold, it does not immediately initiate linear heating. Instead, it uses this moment as the starting point for dynamic discrimination, and continuously acquires images of the area above the sample at high speed during the subsequent preset discrimination period. The core of the algorithm lies in analyzing the dynamic stability of the image texture in this region over time. A normal sublimation process is a molecular-level, gradual, and continuous process. The solid particles formed by recondensation gradually grow or increase. Reflected in the image, the texture change is slow and monotonous, so the temporal variance of the texture time sequence will be very small. Conversely, the boiling of the solvent or the splashing of the sample is an instantaneous and violent physical process. Tiny droplets are ejected, which manifests in the image as abrupt changes in pixel grayscale values, rapid movement, or violent oscillations in shape. This causes violent fluctuations in the texture sequence of the area above, and its temporal variance will increase significantly. Therefore, by calculating this temporal variance and comparing it with a dynamically set threshold based on experience, these two phenomena can be distinguished. If the variance exceeds the limit, a destructive abnormal event is determined to have occurred. The current heating program will be immediately stopped, and the sample will be actively cooled back to the first preset temperature to execute the recovery process.

[0043] In more specific cases, a direct retry might be ineffective because the previous sputtering process may have left solid artifacts (stains) on the observation window, which could be misidentified as new pixels, leading to erroneous triggering. Therefore, the cooling and solidification process of the sputtered material is continuously monitored during cooling. Once image analysis shows that these previously dynamic pixels have stabilized, their shape and position no longer changing, forming solid artifacts, a reference frame for the artifacts is immediately acquired. By comparing this reference frame with the initial, clean baseline image, all newly added, fixed contaminated pixels due to sputtering are located, and a binary artifact mask is generated based on their positional information. Subsequently, real-time image acquisition resumes at a first preset temperature, and new active pixels are identified through inter-frame differencing. All newly detected pixels are compared with the artifact mask; any pixels within the mask's marked area are automatically discarded as historical artifacts. Only newly generated pixels outside the mask, representing true sublimation (i.e., corrected and resampled new pixels), are counted. When the number of these clean pixels exceeds the threshold again, the trigger condition is confirmed, and the subsequent linear heating and data analysis process is initiated.

[0044] The above technical steps proactively diagnose anomalies such as boiling by analyzing the dynamic characteristics of the signal, and stop invalid measurements. The physical artifacts generated by the failure are converted into digital masks, and the interference is ignored in the automatic retry. This enables the measurement process of non-ideal samples to be self-corrected, and improves the robustness and automation of the method.

[0045] S103. When a negative slope range is identified in the main transmittance curve, and the main texture complexity curve shows a non-decreasing trend within the temperature range corresponding to the negative slope range, the transmittance curve data corresponding to the negative slope range shall be excluded in subsequent calculations.

[0046] Among them, the negative slope range refers to a section on the main transmittance curve where the transmittance value decreases as the temperature increases; the non-decreasing trend indicates that the main texture complexity curve remains stable or increases within the same temperature range.

[0047] Specifically, for example, some polycrystalline materials may undergo a solid-solid phase transition before melting, transforming into a denser but less optically transparent crystal form; or some thermotropic liquid crystal materials may form an ordered liquid crystal phase at a specific temperature, whose light scattering ability may be stronger than the original solid phase, both of which lead to a temporary decrease in transmittance. However, these phenomena are not true melting because the material still maintains its solid or liquid crystal structure. In this case, the principal texture complexity curve becomes the key criterion: true melting must be accompanied by the disintegration of the solid structure, i.e., a decrease in texture complexity. If transmittance decreases while texture complexity does not, it is determined to be a non-melting phase transition event. Therefore, through logical judgment, this false signal is identified, and this misleading transmittance data is excluded from the subsequent initial melting point slope calculation, thus avoiding misjudging the non-melting phase transition point as the initial melting point, improving the accuracy of the algorithm and its adaptability to complex samples.

[0048] In some embodiments, the identification and exclusion of this data can be achieved in a variety of ways: Optionally, a logic-based decision method based on event detection is used: peak detection is performed on the main transmittance curve to identify all local maxima; for each descending segment after a maxima, the corresponding texture complexity curve segment is analyzed; if the overall trend of the texture complexity curve segment is stable or rising (which can be determined by the slope of linear regression), then the data of the entire transmittance descending segment is logically masked or assigned zero weight in subsequent calculations.

[0049] Understandably, more complex signal processing techniques, such as wavelet transform, can be used to identify transient anomalies and combined with logical rules to exclude data; this is not a limitation here.

[0050] In some embodiments, if the change in the optical properties of the sample to be tested is not significant enough to cause the slope of the main transmittance curve to fail to meet the preset threshold, the method can be switched to an alternative criterion to determine the initial melting point of the sample based solely on the main texture complexity curve, thereby enhancing the method's adaptability to special materials (such as dark or slow-melting samples) and its detection success rate.

[0051] Specifically, this step exists as a supplement and error-tolerant mechanism to the standard initial melting point criterion (S104). Under certain circumstances, the melting process of the sample may not be accompanied by a significant increase in transmittance. The physical mechanisms are as follows: First, for dark-colored or inherently opaque samples, even after melting from a solid powder to a liquid, the liquid itself still has an extremely high light absorption rate, resulting in almost no increase in the intensity of transmitted light, and the main transmittance curve appears as a flat line without a clear upward trend; Second, for certain polymers or high-viscosity substances, the melting process may be a very slow softening process over a wide temperature range, unlike the rapid phase transition of amorphous substances. This results in an extremely gradual increase in transmittance, and the slope of the curve may be less than the preset first positive threshold at any given time. In these cases, the S104 step, which strictly relies on both transmittance and texture complexity verification, will fail to find any temperature point that meets the conditions, leading to measurement failure.

[0052] During the S104 search, if a first melting point meeting both conditions is not found even at a higher temperature (e.g., exceeding the expected melting point range), this backup logic is activated. This logic temporarily ignores the slope condition of the main transmittance curve and instead focuses on analyzing the main texture complexity curve. Because regardless of the sample's color intensity or melting speed, the transformation from an irregular collection of solid particles to a homogeneous liquid phase leads to the collapse and homogenization of the sample region's macroscopic physical structure, a process that is consistently reflected in the continuous decrease of main texture complexity. The main texture complexity curve is rescanned from a low temperature to find the inflection point where it transitions from a stable or fluctuating state to a sustained unidirectional downward trend. This inflection point marks the moment when the sample's macroscopic solid structure begins to disintegrate; although optical signals cannot confirm this, the change in physical structure has clearly occurred. Therefore, the temperature corresponding to this inflection point is determined as the initial melting point of the structure.

[0053] The aforementioned technical steps enhance the method's versatility. Since the melting of matter from a solid to a liquid state results in a more fundamental and universal characteristic of macroscopic physical structure breakdown (reduced texture complexity) than changes in optical properties (increased transmittance). Therefore, when the optical signal is weak due to the sample's inherent properties (such as its dark color), the method relies on a more reliable structural signal, avoiding measurement failures and expanding the instrument's applicability.

[0054] S104. The temperature at which the slope of the main transmittance curve first exceeds the preset first positive threshold and the main texture complexity curve simultaneously begins to decrease is determined as the initial melting point of the sample.

[0055] The preset first positive threshold is the minimum slope value of the transmittance curve used to define the significant start of melting. This value is set empirically through statistical analysis of melting curves of a large number of different substances, aiming to effectively filter out baseline noise fluctuations and ensure that the starting point of the melting process is captured.

[0056] Specifically, the initial melting point, or the initial melting point, is physically defined as the temperature at which the first crystal begins to melt. In macroscopic imaging signals, this microscopic event manifests as a synergistic and irreversible transformation of the sample's optical properties and physical structure. Solely relying on an increase in transmittance may be affected by factors such as changes in sample color; solely relying on a decrease in texture complexity may be affected by slight sample movement or collapse. This method requires that the slope of the transmittance curve must first exceed a clear and significant threshold (representing the effective increase in optical transparency), and at the same moment, the texture complexity curve must also begin to decrease (representing the actual collapse of the solid particle structure). Coupled with these two independent physical observation dimensions, a highly reliable event trigger is formed, which searches along the temperature axis to find the first temperature point that simultaneously satisfies both conditions and designates it as the initial melting point of the sample.

[0057] In some embodiments, the initial melting point can be determined in a variety of ways: Optionally, a dual-threshold synchronous scanning method is used: calculate the first derivative (slope) sequence of the main transmittance curve and the main texture complexity curve; scan point by point starting from low temperature to find the first temperature point T that satisfies the condition that the transmittance slope [T] > the first positive threshold and the texture complexity slope [T] < the negative threshold (e.g., -0.02, indicating a clear decrease); directly determine this temperature point T as the initial melting point.

[0058] Understandably, a machine learning-based classifier can also be used to identify curve patterns that match the characteristics of the initial melting point, thereby determining the initial melting point; this is not a limitation here.

[0059] S105. The temperature at which the slope of the main transmittance curve approaches zero and the main texture complexity curve stabilizes at the lowest value plateau is determined as the final melting point of the sample.

[0060] Among them, a slope approaching zero means that the slope of the main transmittance curve is consistently lower than a preset minimum value within a small temperature range, indicating that the transmittance has reached its maximum and no longer changes with increasing temperature; stabilizing at the lowest value plateau indicates that the main texture complexity curve has reached the trough of the value and remains stable, indicating that the sample has been completely liquefied and the internal structure no longer changes.

[0061] Specifically, the endpoint of the melting process, i.e., the final melting point, is physically defined as the temperature at which the last solid crystal completely melts. Similar to determining the initial melting point, a principle of dual evidence is employed to ensure the reliability of the results. When the melting process ends, the sample completely transforms into a homogeneous liquid phase, and its physical and optical properties reach a new stable state. First, the transmittance reaches its maximum value and then stops increasing; the slope of the curve changes from a positive value to near zero, forming a plateau or asymptote, marking the end of optical changes. Second, the texture complexity continuously decreases during the melting process, reaching its lowest value corresponding to pure liquids, and then stops changing, also forming a flat valley plateau, marking the end of macroscopic structural changes. By setting a very small slope threshold and a minimum stability duration window, it is detected whether both curves have entered their respective stable plateau periods. The starting temperature point where both curves have entered stable plateaus is identified as the final melting point of the sample.

[0062] In some embodiments, the final melting point can be determined in a variety of ways: Optionally, the sliding window variance method is used: after the initial melting point, the numerical variance of the main transmittance curve and the main texture complexity curve within the window is calculated using a sliding window; the first temperature point T is found such that within a continuous window starting from T, the variance of both the transmittance curve and the texture complexity curve are continuously less than a preset stable threshold; this temperature point T is determined as the final melting point.

[0063] Understandably, finding the zero-crossing points of the second derivatives of the two curves can also help determine the inflection point and the flat region, thus more accurately locating the final melting point; however, this is not a limitation here.

[0064] To more intuitively illustrate the melting point determination method of this embodiment, please refer to [link to relevant documentation]. Figure 2 . Figure 2 The diagram shows the principal transmittance and principal texture complexity curves during the sample melting process. It illustrates how the two core data curves generated by this method, principal transmittance curve A and principal texture complexity curve B, change with temperature during a typical melting point measurement.

[0065] The horizontal axis represents the real-time temperature of the sample area, the left vertical axis represents the main transmittance of the sample area, and the right vertical axis represents the main texture complexity.

[0066] like Figure 2As shown, in the initial stage of heating, the sample is in a solid state, with the main transmittance curve A remaining at a low level, while its powder structure is complex, thus the main texture complexity curve B remains at a high level. When the temperature reaches the initial melting point, two key synergistic changes are detected simultaneously: the slope of the main transmittance curve A exceeds the preset first positive threshold for the first time, indicating that the sample begins to melt and become transparent; at the same time, the main texture complexity curve B also begins to show a continuous downward trend from this moment, which physically verifies the collapse and liquefaction of solid particles. Therefore, this temperature point is determined as the initial melting point.

[0067] During further heating, anomalies such as those observed in the negative slope interval C may occur, such as the formation and rupture of bubbles within the sample, leading to a sudden drop in transmittance. In this case, our method, through simultaneous analysis of the main texture complexity curve B, reveals that the curve exhibits a plateau-like, non-decreasing trend within this interval. This indicates that the phenomenon is not due to sample solidification, but rather other physical perturbations. Therefore, transmittance data within the negative slope interval C are excluded, thus avoiding interference with subsequent calculations.

[0068] Finally, when the temperature reaches the final melting point, the sample completely melts into a homogeneous liquid. At this point, the slope of the principal transmittance curve A approaches zero, entering a saturated high-value plateau, indicating that transmittance no longer increases; simultaneously, the principal texture complexity curve B also stabilizes at its lowest value plateau, indicating that the internal structure of the sample has reached its most homogeneous state. Based on the simultaneous satisfaction of these two conditions, this temperature point is determined as the final melting point of the sample.

[0069] In the above embodiments, by introducing the principal texture complexity curve and the principal transmittance curve for collaborative analysis, when it is identified that the transmittance decreases in a non-true manner due to solid-phase transformations such as sample collapse (i.e., the principal transmittance curve has a negative slope range), the key feature that the principal texture complexity curve shows a non-decreasing trend at this time is simultaneously judged. This identifies and eliminates artifact data, ensuring that the subsequent determination of the initial melting point is based on the signal changes in the true melting stage, that is, the moment when the transmittance first increases significantly and the texture complexity begins to decrease simultaneously. This solves the technical defect in related technologies that misjudge the initial melting point due to changes in the physical morphology of the sample, and improves the accuracy and reliability of melting point determination.

[0070] In other embodiments of this application, when complex physical changes such as softening and decomposition occur during sample heating, these artifacts may be misjudged as the initial melting point due to single-dimensional signal abrupt changes. Using the video melting point acquisition method provided in this application, preliminary screening can be performed by fusing features from two dimensions: transmittance and texture complexity. Furthermore, the stability and persistence of the melting trend can be double-checked within a subsequent confirmation window, thereby eliminating interference.

[0071] like Figure 3The diagram shown is another flowchart illustrating the video melting point acquisition method provided in this application embodiment, including the following steps: S301. Raise the sample area from the initial temperature to the first preset temperature and continuously acquire real-time images. By performing a difference operation between the real-time images and the reference images, identify the newly generated pixels above the sample area.

[0072] S302. When the number of new pixels exceeds the preset pixel threshold, during the linear heating process of the sample area, image data analysis is performed on the sample area to generate the main transmittance curve and the main texture complexity curve.

[0073] Steps S301-S302 and Figure 1 Steps S101-S102 in the illustrated embodiment are similar and can be found in the descriptions of steps S101-S102, which will not be repeated here.

[0074] In some embodiments, when the sample as a whole exhibits a slowly changing, low-activity melting state, the detection accuracy of samples with complex melting behavior can be improved by reducing global monitoring to multiple independent sub-regions and capturing non-uniform or weak initial melting signals.

[0075] During standard linear heating, some samples (such as polymers, crystals containing impurities, or samples with uneven packing) may not exhibit vigorous, synchronous melting. Instead, they may melt slowly starting from one or a few small regions. In this case, the globally calculated principal transmittance curve and principal texture complexity curve, being averages of the signal across the entire sample region, will show very weak and gradual changes. These changes are unlikely to trigger the initial melting point determination based on significant slope changes in S304, thus constituting a low-activity melting state.

[0076] Specifically, the status of the master curve is continuously monitored. A low-activity melting state is defined as simultaneously meeting the following conditions: the slope of the master transmittance curve remains positive (indicating a weak transmittance trend), but the value is consistently below the first positive threshold (this slope value is empirically set based on extensive experimental data from typical materials melting, used to distinguish significant melting from background fluctuations); simultaneously, the slope of the master texture complexity curve remains negative (indicating weak structural collapse), but the absolute value is less than a preset internal change threshold (this threshold is set based on statistical analysis of the noise level of image textures under constant temperature conditions, used to filter out meaningless noise fluctuations). When the duration of this low-activity state or the corresponding temperature rise exceeds a preset temperature / time window threshold (this window threshold is a judgment period set empirically, such as lasting 5 seconds or a temperature rise exceeding 2°C, used to confirm that the state is a continuous trend rather than a transient disturbance), the global analysis method is deemed to have failed, and a local refined analysis mode is switched.

[0077] In this mode, the entire sample area image is first logically divided into a grid, such as a 3x3 or 4x4 checkerboard, forming multiple independent sub-regions. Then, during the subsequent heating process, instead of simply calculating the global transmittance and texture complexity, the image data of each sub-region is analyzed independently and in parallel, generating a unique sub-transmittance curve and sub-texture complexity curve for each sub-region. If melting occurs only in one or two sub-regions, this localized, drastic physical change will produce a very significant signal inflection point on the corresponding sub-curve, without being diluted by data from other unmelted areas. Finally, the judgment criteria in S304 are applied to each sub-region: once the slope of the sub-transmittance curve corresponding to any sub-region first exceeds a preset sub-region positive threshold (this threshold is specifically for sub-region detection, and its value is empirically adjusted based on the ratio of the sub-region area to the total area), and the sub-texture complexity curve also begins to decrease simultaneously, the temperature at this moment is determined as the true initial melting point of the sample. This is because, physically speaking, the melting of any part of the sample signifies that initial melting has occurred.

[0078] The above technical steps can overcome the limitations of global averaging analysis. For non-uniformly molten samples, initial melting often begins in local micro-regions, and the signal is submerged by averaging of the unmelted areas on the global curve, leading to detection delays and inaccuracies. By reducing the dimensionality of the analysis unit to sub-regions, the local melting signal can be amplified, forming clearly identifiable characteristic inflection points on the sub-curve, thereby capturing the occurrence of real physical events in real time and improving the sensitivity and accuracy of detection.

[0079] S303. When a negative slope range is identified in the main transmittance curve, and the main texture complexity curve shows a non-decreasing trend within the temperature range corresponding to the negative slope range, the transmittance curve data corresponding to the negative slope range shall be excluded in subsequent calculations.

[0080] Step S303 and Figure 1 Step S103 in the illustrated embodiment is similar and can be found in the description of step S103, which will not be repeated here.

[0081] S304. On the main transmittance curve and the main texture complexity curve, take all temperature points where the slope of the transmittance curve exceeds the first positive threshold and the texture complexity curve begins to decrease as candidate points for initial melting.

[0082] Among them, the texture complexity curve begins to decline when the first derivative (i.e., the slope) of the data curve first changes from a non-negative value to a negative value; the initial melting candidate point refers to a temporary temperature marker point on the data curve that initially meets the melting initiation characteristics but still needs further verification.

[0083] Specifically, after obtaining the principal transmittance and principal texture complexity curves, the two curves are traversed point by point or through a small sliding window, and a dual-condition simultaneous judgment is performed at each temperature point. The first condition is that the slope of the transmittance curve exceeds a first positive threshold, which physically corresponds to the sample beginning to transform from an opaque solid powder into a transparent liquid phase, and this transformation rate is significant, thus excluding weak transmittance changes caused by instrument noise or non-melting factors such as thermal expansion of the sample. The second condition is that the texture complexity curve begins to decrease, which physically corresponds to the initial moment when the sample's inherent, irregular particle structure begins to collapse and fuse, transitioning into a homogeneous liquid phase. By combining these two conditions through logical AND operations, temperature points that simultaneously exhibit clear melting signs in both optical properties and macroscopic structure can be screened out. These points are marked as initial melting candidate points and stored in a set.

[0084] In some embodiments, the screening of initial melting candidate points can be achieved in a variety of ways: Optionally, a point-by-point differentiation method is used: Forward difference or central difference is applied to the discrete main transmittance curve and main texture complexity curve data points to calculate the instantaneous slope corresponding to each temperature point; all temperature points are traversed, and it is checked whether the transmittance slope of the current point i is greater than the first positive threshold; at the same time, it is checked whether the texture complexity slope of the current point i is negative and the slope of its previous point i-1 is non-negative, so as to accurately lock the inflection point where the decline begins; if a temperature point satisfies the above two conditions at the same time, it is recorded as a candidate point for initial melting.

[0085] It is understandable that other methods such as sliding window fitting and wavelet transform can also be used to achieve this step, and no limitation is made here.

[0086] S305. Starting from each initial melting candidate point, a preset continuous temperature zone is set as a confirmation window to check the steady drop state that appears within the confirmation window.

[0087] The sustained temperature zone refers to a fixed temperature range starting from the candidate point temperature, also known as the confirmation window. Its width (e.g., 5℃ or 10℃) is set empirically based on a large number of melting experiments, covering a typical temperature rise range sufficient to observe a stable melting trend. The stable decreasing state is a logical state describing the stability of the decreasing trend of texture complexity, in which the slope of the main texture complexity curve is consistently negative or zero and no positive slope exceeding the preset fluctuation threshold appears. The preset fluctuation threshold is a small positive slope value close to zero, a fault tolerance value set based on the assessment of the system measurement noise level, used to allow the curve to have extremely weak local rises caused by noise in the overall decreasing trend, thereby avoiding misjudgment.

[0088] Specifically, each candidate point selected by S304 is verified to confirm whether the structural collapse indicated by the candidate point is a continuous and irreversible process, rather than a transient, recoverable disturbance or noise.

[0089] For each initial melting candidate point in the list, a confirmation window [T_cand, T_cand+ΔT_window] is defined starting from the temperature T_cand corresponding to that point and moving towards higher temperatures. Then, the slope of the main texture complexity curve is examined for all data points within this window. In an ideal melting process, structural collapse should be continuous, therefore its slope should always be negative. However, considering the unavoidable random noise in actual measurements, strictly requiring a constantly negative slope may be too demanding. Therefore, a steady-state definition is introduced: within the entire confirmation window, the slope of the curve can be negative or zero (representing a decrease or a temporary halt in the decrease), and even extremely small positive slopes are allowed, but the magnitude of this positive slope must never exceed a preset fluctuation threshold. If the slope of the curve consistently meets this condition throughout the entire window, the candidate point is considered to have passed the steady-state test, proving that the structural change it induces is stable and continuous. Conversely, if any point within the window has a slope exceeding the fluctuation threshold, it indicates that the structure may have undergone reorganization or that the measurement has been significantly disturbed. This candidate point is considered a false melting point signal and will be removed from the candidate list.

[0090] In some embodiments, the verification of the steady-state condition can be achieved in a variety of ways: Optionally, a point-by-point traversal and comparison method is adopted: For a candidate point of initial melting, the start and end data indices of its confirmation window are determined; a loop is used to traverse each data point in the confirmation window; in the loop, the instantaneous slope of the texture complexity of the current point is calculated and compared with a preset fluctuation threshold; if the slope of any point is found to be greater than the threshold, the loop is immediately terminated, the candidate point is determined not to meet the steady-state condition, and the next candidate point is checked; if the loop ends normally, the candidate point is determined to meet the steady-state condition.

[0091] It is understandable that other methods, such as setting the proportion of positive slope points within a window or slope variance analysis, can also be used to verify this steady-state condition, and no limitation is made here.

[0092] S306. Verify the steady-increase state within the confirmation window.

[0093] Among them, the steady growth state is a logical state used to describe the health of the light transmittance growth trend, which is that the slope of the main light transmittance curve is continuously positive and the acceleration of growth does not show a continuous negative situation; the confirmation window is exactly the same as the continuous temperature zone defined and used in S305.

[0094] Specifically, for candidate points confirmed to meet the steady-state melting condition, the morphology of the main transmittance curve is analyzed within the same confirmation window. First, the slope of the curve must remain consistently positive within this window to ensure that the optical transparency of the sample is steadily increasing, a fundamental optical characteristic of melting. Second, an analysis of acceleration, i.e., the trend of the slope, is introduced. In a real initial melting process, the melting rate is usually stable or even accelerating in the early stages. If the transmittance curve quickly exhibits a sustained negative acceleration after the initial melting point (i.e., the curve shape changes from rising to flat), this may indicate that the melting process is very short-lived, unstable, or has encountered an obstacle, or even just an artifact. Therefore, the second derivative of the curve is calculated. If the second derivative is found to be negative for a continuous interval within the confirmation window, it is determined that the steady-state melting condition is not met. Only when the transmittance curve simultaneously meets both the conditions of a positive slope and non-sustained negative acceleration within this window is the optical change induced by the candidate point considered to conform to the ideal melting model, thus passing this test.

[0095] In some embodiments, the verification of a steady-growth state can be achieved in multiple ways: Optionally, a local polynomial fitting method can be used: for a candidate point and its confirmation window, extract the transmittance data points within the window; perform a quadratic or cubic polynomial fitting on these data points to obtain a result of the form y=ax 2 +bx+c or y=ax 3 +bx 2 Find the fitting function for +cx+d; analyze the coefficients of the fitting function. For quadratic fitting, the first derivative 2ax+b must be consistently positive within the window, and the second derivative 2a must not be significantly negative. Based on the coefficient analysis results, directly determine whether the entire window satisfies the steady-increase state. For example, if a is significantly negative, it indicates the existence of continuous negative acceleration, and the condition is determined not to be met.

[0096] It is understandable that linearity can also be analyzed after logarithmic transformation, or the steady-increase state can be verified by model matching, which is not limited here.

[0097] S307. The first initial melting point that simultaneously satisfies both the steady-fall state and the steady-rise state is ultimately determined as the initial melting point of the sample.

[0098] Among them, "first" refers to the point with the lowest corresponding temperature value among all candidate points that pass the test; "simultaneously satisfying" means that the candidate point passes both the steady-fall state test in S305 and the steady-increase state test in S306.

[0099] Specifically, following the aforementioned steps, one or more candidate points that passed both stability and persistence verification were obtained. According to the physical definition of melting point, the initial melting point is the initial temperature at which a substance begins to melt. Therefore, among all points that meet the conditions, the one with the lowest temperature is selected as the final result. All candidate points that passed both S305 and S306 tests are sorted in ascending order of temperature value, and then the first candidate point (i.e., the one with the lowest temperature) is directly selected, and its corresponding temperature value is output as the initial melting point of the sample.

[0100] In some embodiments, after the initial melting point is determined, a verification window can be introduced to identify pseudo-melting phenomena such as recrystallization, aiming to eliminate interference from complex physical processes such as polymorphic transformation, thereby improving the accuracy of the final melting point determination.

[0101] For special samples exhibiting complex phase transition behaviors (such as polymorphism and isomorphism), the first temperature point in S307 that physically matches the ideal melting model (transmittance, texture complexity) was found. However, this point may not be the final, stable melting point of the sample, but rather the melting point of a metastable crystalline form. Many substances exist in different crystalline forms, among which metastable crystalline forms have lower melting points. The liquid phase formed after melting may recrystallize to form a more stable crystalline form with a higher melting point as the temperature continues to rise; this process is called recrystallization. Without proper differentiation, the instrument may incorrectly report this metastable melting point as the final result.

[0102] Specifically, the first candidate initial melting point identified in S307 that simultaneously satisfies both steady-fall and steady-rise conditions is temporarily marked as a pending initial melting point, rather than immediately used as the final result. Subsequently, a verification window is defined immediately after this pending initial melting point. This window is a preset verification temperature range, the width of which (e.g., 5-15℃) is empirically set based on extensive experiments on substances known to have polymorphic transformation characteristics, aiming to provide a sufficient temperature rise range to observe whether the melted phase is stable.

[0103] Within this verification window, the main transmittance curve and the main texture complexity curve are continuously monitored to actively search for recrystallization features. Recrystallization is the reverse process of melting, so its performance in image data is completely opposite to melting: the slope of the main texture complexity curve changes abruptly from negative or zero to positive, which physically corresponds to the sudden generation of new, irregular solid crystal nuclei in a homogeneous liquid phase, resulting in a more complex macroscopic structure; or, the slope of the main transmittance curve changes abruptly from positive to negative, which corresponds to the precipitation of opaque solid particles in a transparent liquid phase, resulting in a decrease in overall transmittance.

[0104] If no recrystallization characteristic is detected within the entire verification window, it proves that the melting caused by the undetermined initial melting point is stable and irreversible, and the undetermined initial melting point is ultimately determined as the initial melting point of the sample. Conversely, if a clear recrystallization characteristic is detected within the verification window, the previously undetermined initial melting point is determined to be a pseudo-melting point (melting point of a metastable crystal form) and is immediately excluded. This candidate point is discarded, and starting from the temperature point after that point, the logic returns to S304 to continue searching for the next initial melting candidate point that meets the conditions and uses it as the new undetermined initial melting point. The above verification process is repeated until a melting point that passes the recrystallization verification is found, or until all candidate points have been traversed.

[0105] The above steps distinguish between thermodynamically reversible phase transitions and irreversible melting. By adding a verification window after the initial melting candidate point and actively searching for recrystallization signals that are opposite to the melting characteristics, false melting points caused by the melting of metastable crystal forms and recrystallization of stable crystal forms can be eliminated. Thus, the true melting point of the sample's stable crystal form is ultimately determined, improving the reliability of the method.

[0106] S308. The temperature at which the slope of the main transmittance curve approaches zero and the main texture complexity curve stabilizes at the lowest value plateau is determined as the final melting point of the sample.

[0107] Step S308 and Figure 1 Step S105 in the illustrated embodiment is similar and can be found in the description of step S105, which will not be repeated here.

[0108] In the above embodiments, the dual features of transmittance and texture complexity are used to screen out candidate melting points. Then, a confirmation window mechanism is introduced. By checking whether the melting trend meets the continuous conditions of steady decrease and steady increase within the window, the judgment criteria are upgraded from a single instantaneous inflection point to a stable process. This can eliminate artifact interference such as sample softening, cracking, or trace decomposition that can only cause instantaneous signal changes. This ensures that the final identified melting point is a real and irreversible physical phase transition point, thereby improving the anti-interference ability of the detection and the accuracy of the final result.

[0109] The exemplary melting point apparatus 400 provided in the embodiments of this application is described below. Figure 4 This is an exemplary hardware structure diagram of the melting point apparatus 400 provided in the embodiments of this application.

[0110] In some embodiments, the melting point apparatus 400 is a computer device or includes a computer device. The computer device includes a processor, memory, and a network interface connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores data. The network interface of the computer device is used to communicate with other external terminals or servers via a network connection. In some embodiments, the network interface can be a wired network interface; in some embodiments, the network interface can also be a wireless network interface. When the computer program is executed by the processor, it implements the methods described in the embodiments of this application.

[0111] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0112] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0113] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0114] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0115] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for video melting point acquisition, characterized in that, The method comprises the following steps: The sample region is raised from an initial temperature to a first preset temperature, and real-time images are continuously collected. New pixel points newly generated above the sample region are identified by performing a differential operation on the real-time images and a reference image. The reference image is an image containing a sample to be tested captured at the initial temperature. The sample region is a sample powder initial contour region determined according to the reference image. In a case where the number of new pixel points exceeds a preset pixel point threshold, image data analysis is performed on the sample region during a linear temperature rising process of the sample region, and a main transmittance curve and a main texture complexity curve are generated. The main texture complexity curve is used to represent the change degree of pixel gray value in the sample region. When it is identified that the main transmittance curve has a negative slope interval and the main texture complexity curve presents a non-decreasing trend within a temperature range corresponding to the negative slope interval, transmittance curve data corresponding to the negative slope interval is excluded in subsequent calculation. A temperature at which the slope of the main transmittance curve first exceeds a preset first positive threshold and the main texture complexity curve starts to decrease at the same time is determined as an initial melting point of the sample. A temperature at which the slope of the main transmittance curve approaches zero and the main texture complexity curve stabilizes at a lowest value platform is determined as a final melting point of the sample.

2. The method of claim 1, wherein, The temperature at which the slope of the main transmittance curve first exceeds the preset first positive threshold and the main texture complexity curve starts to decrease at the same time is determined as the initial melting point of the sample, and specifically comprises the following steps: On the main transmittance curve and the main texture complexity curve, temperature points at which the slope of the transmittance curve exceeds the first positive threshold and the texture complexity curve starts to decrease are taken as initial melting candidate points. A preset continuous temperature range is set as a confirmation window backward from each initial melting candidate point as a starting point, and a stable decrease state appearing in the confirmation window is verified. The stable decrease state is that the slope of the main texture complexity curve is continuously negative or zero and no positive slope exceeding a preset fluctuation threshold appears. A stable increase state in the confirmation window is verified. The stable increase state is that the slope of the main transmittance curve is continuously positive and the acceleration of growth does not continuously become negative. The initial melting candidate point first satisfying the stable decrease state and the stable increase state is finally determined as the initial melting point of the sample.

3. The method of claim 2, wherein, The initial melting candidate point first satisfying the stable decrease state and the stable increase state is finally determined as the initial melting point of the sample, and specifically comprises the following steps: The initial melting candidate point first satisfying the stable decrease state and the stable increase state is determined as a tentative initial melting point. In a case where a recrystallization feature is detected in a verification window, the tentative initial melting point is excluded, and a next initial melting candidate point after the tentative initial melting point is taken as a tentative initial melting point. The verification window is a preset verification temperature range set after the confirmation window. The recrystallization feature is that the slope of the main texture complexity curve suddenly changes from negative or zero to positive, or the slope of the main transmittance curve suddenly changes from positive to negative. In the case where no recrystallization feature is detected within the verification window, the pending preliminary melting point is finally determined as the preliminary melting point of the sample.

4. The method of claim 1, wherein, In the case where the number of new pixel points exceeds the preset pixel point threshold, before generating the main transmittance curve and the main texture complexity curve through image data analysis on the sample region during the linear temperature rising process, the method further comprises: Taking the time when the new pixel point is detected as a dynamic discrimination starting point, continuously performing image data analysis on the upper region where the new pixel point is located within a preset discrimination time period after the dynamic discrimination starting point to generate an upper region texture time sequence; Calculating a time domain variance of the upper region texture time sequence; When the time domain variance exceeds a preset dynamic threshold, stopping the current linear temperature rising process and recooling the temperature of the sample region to the first preset temperature; At the first preset temperature, re-starting continuous acquisition of resampling real-time images of the sample region, and identifying newly generated resampling new pixel points of the sample region through inter-frame difference operation on the resampling real-time images, and using the resampling new pixel points to replace the new pixel points.

5. The method of claim 4, wherein, The re-starting continuous acquisition of resampling real-time images of the sample region at the first preset temperature, and identifying newly generated resampling new pixel points of the sample region through inter-frame difference operation on the resampling real-time images, and using the resampling new pixel points to replace the new pixel points, specifically comprises: Continuously monitoring images above the sample region, when a dynamic droplet pixel point is changed into a stable pixel point and is stable at the first preset temperature, acquiring an image containing a solid-state artifact as an artifact reference frame, and comparing the artifact reference frame with the reference image to identify difference pixel points, and generating an artifact mask based on the positions of the difference pixel points; Re-starting continuous acquisition of resampling real-time images of the sample region at the first preset temperature, and identifying newly generated resampling new pixel points of the sample region through inter-frame difference operation on the resampling real-time images; Removing pixel points in the resampling new pixel points that coincide with the mask mark positions in space to obtain corrected resampling new pixel points, and using the corrected resampling new pixel points to replace the new pixel points.

6. The method of claim 1, wherein, In the case where the main texture complexity curve presents a non-decreasing trend within a temperature range corresponding to the negative slope interval when the negative slope interval of the main transmittance curve is identified, after excluding transmittance curve data corresponding to the negative slope interval in subsequent calculation, the method further comprises: When it is detected that the main texture complexity curve presents a continuous decreasing trend and the slope of the main transmittance curve in the corresponding temperature interval fails to satisfy the first positive threshold condition, determining the structural preliminary melting point of the sample only according to the temperature corresponding to the starting time of the continuous decrease of the main texture complexity curve.

7. The method of claim 1, wherein, In the case where the main transmittance curve and the main texture complexity curve are generated through image data analysis on the sample region, the method further comprises: In the linear temperature rising process, when the low activity melting state lasts for a preset temperature window threshold or a preset time window threshold, the sample region is divided into a grid containing multiple independent sub-regions; the low activity melting state is that the slope of the main transmittance curve is continuously positive, the value of the main transmittance curve is lower than the first positive threshold, or the slope of the main texture complexity curve is continuously negative, and the absolute value of the main texture complexity curve is less than a preset internal change threshold; Calculate the image data of each sub-region, and generate a corresponding sub-transmittance curve and a sub-texture complexity curve for each sub-region; Determine the temperature when the slope of the corresponding sub-transmittance curve of each sub-region first exceeds the preset sub-region positive threshold and the corresponding sub-texture complexity curve starts to decrease at the same time as the real initial melting point.

8. A melting point apparatus characterized by, The melting point instrument comprises one or more processors and a memory; the memory is coupled with the one or more processors, and the memory is used to store computer program code, the computer program code comprises computer instructions, and the one or more processors invoke the computer instructions to enable the melting point instrument to perform the method according to any one of claims 1-7.

9. A computer program product comprising instructions, characterized in that, When the computer program product is running on the melting point instrument, the melting point instrument is enabled to perform the method according to any one of claims 1-7.

10. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are running on the melting point instrument, the melting point instrument is enabled to perform the method according to any one of claims 1-7.

Images