Low-freezing-point infiltration and visualization integrated detection method for improved salinized soil in seasonal frozen region

Abstract

The invention relates to a low-freezing-point infiltration and visualization integrated detection method for improved saline soil in a seasonal frozen region. Cooperative improvement is realized around three links of low-temperature, constant-temperature and constant-flow liquid supply, in-situ visualization imaging and synchronous control and quantitative identification. The method comprises the following steps: stably supplying a tracing solution with the freezing point lower than 0 DEG C in a range of-5 DEG C to + 5 DEG C; the continuous wetting and infiltration process of the surface layer of the soil sample is kept uninterrupted; a non-destructive in-situ observation window is constructed through fixed geometry and exposure combination of a wedge-shaped or cutting edge transparent observation module, an industrial camera and annular illumination; the central control and data processing module is used for implementing timestamp alignment and closed-loop coordination on pump speed, temperature control, illumination and imaging; according to the system, interface and alignment errors are reduced, energy consumption and arrangement complexity are reduced, repeatability, comparability and efficiency of tests are improved, and a standardized data basis is provided for mechanism research, parameter inversion and model calibration under the condition of an indoor rack / transparent soil column.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of soil hydrology monitoring, in particular to a low-ice-point infiltration and visualization integrated detection method for improved saline soil in the seasonal frozen region. BACKGROUND

[0002] In the seasonal frozen region, the saline soil is prone to form a fissure-macro-pore structure in the freezing and thawing cycle, which induces preferential flow and affects the water and solute migration law. In order to strip the environmental uncertainty and improve the repeatability, more and more related researches are carried out under indoor controlled conditions. However, there are mainly the following problems in the existing experimental process: (1) Infiltration or drip irrigation under normal temperature bench is prone to local icing and air resistance near 0℃, resulting in a decrease in flux control accuracy; (2) The destructive sampling required by the dissection type dyeing and the later slice imaging makes it difficult to realize in-situ and continuous observation; (3) The indirect criterion relying only on the conductivity / humidity response has limited spatial resolution, and it is difficult to correspond the flux to the specific fissure-macro-pore structure; (4) The indoor device is mostly split (liquid supply, cooling, imaging and analysis are scattered), and the temperature, humidity and light are not controlled synchronously, resulting in insufficient repeatability and comparability.

[0003] Therefore, an integrated bench-type monitoring system is urgently needed under the indoor test conditions, which integrates low-ice-point constant-temperature and constant-flow liquid supply, in-situ visualization of transparent window industrial camera and synchronous control and image recognition processing. It can stably operate in the range of -5~+5℃, output the preferential flow geometry and statistical parameters (such as area ratio, path width, depth, fractal characteristics, etc.), so as to meet the experimental requirements of high repeatability and quantifiable parameters in the laboratory. SUMMARY

[0004] The technical problem solved by the present application is to provide a low-ice-point infiltration and visualization integrated detection method for improved saline soil in the seasonal frozen region, which can solve the problems in the background.

[0005] To solve the above technical problems, the technical scheme of the present application is as follows: a low-ice-point infiltration and visualization integrated detection method for improved saline soil in the seasonal frozen region, the innovation point of which is as follows: S1: Module preparation: A low-temperature constant-flow liquid supply module is set up for supplying constant-temperature and constant-flow tracer solution with an ice point below 0℃ and sending the tracer solution to the surface layer of the soil sample. The low-temperature constant-flow liquid supply module includes a refrigeration insulation box, a liquid storage tank, a peristaltic pump, a flow meter and a liquid application assembly. The output end of the liquid storage tank is connected to the peristaltic pump, the flow meter and the liquid application assembly through an insulation hose in sequence, and the tracer solution output from the output end of the liquid application assembly enters the soil for natural infiltration; An in-situ image acquisition module is arranged for non-destructive in-situ continuous imaging of the preferential flow; the in-situ image acquisition module comprises a transparent observation module, an industrial camera and an LED array light source; A central control and data processing module is arranged to be connected with the low-temperature constant-flow liquid supply module and the in-situ image acquisition module, for coordinating the timing of liquid supply, lighting and imaging, adjusting the stability of liquid supply based on flow measurement, and processing the collected images to obtain the preferential flow characterization parameters; S2: Pre-cooling and constant temperature: start the refrigeration incubator to the set temperature, and pre-cool the tracer liquid in the liquid storage tank to the target temperature; S3: Flow setting and checking: set the target flow of the peristaltic pump by the central control and data processing module, and check and record according to the flow meter reading; S4: Liquid application and natural infiltration: introduce the incubation hose to the ground liquid application position, start the liquid supply, and the tracer liquid forms a stable thin layer and naturally infiltrates; S5: Synchronous imaging and lighting: control the industrial camera and LED array light source to work synchronously at the set interval by the central control and data processing module, and obtain continuous images at the transparent observation module.

[0006] S6: Data processing and quantitative analysis: perform denoising, brightness normalization and threshold segmentation processing in the central control and data processing module, output the area ratio, maximum / average infiltration depth and dominant path width indicators, generate time series curves and reports.

[0007] Further, the refrigeration incubator in S1 adopts a double-layer insulation structure, and is internally provided with a semiconductor refrigeration and temperature sensor, so that the environment in the incubator is stably controlled at a set low temperature of -2℃±0.5℃, and the tracer liquid is ensured not to freeze at low temperature.

[0008] Further, the tracer solution in the liquid storage tank in S1 is an ethylene glycol / propylene glycol-water system, and visual dye is added; the liquid storage tank has the requirements of low-temperature resistance and corrosion resistance, and is provided with a liquid level mark and a quick connection interface.

[0009] Further, the peristaltic pump in S1 provides adjustable constant flow; the pump tube is made of low-temperature resistant material to adapt to the tracer liquid.

[0010] Further, the flow meter in S1 is electromagnetic or micro-turbine type, which is used for monitoring instantaneous flow and cumulative amount and recording.

[0011] Further, a plurality of liquid application positions are arranged on the soil surface in S1, and a liquid application diffusion head / drip irrigation head can be optionally installed on the liquid application assembly to improve the uniformity of liquid application, and 3-5mm thin layer of fine sand can be laid around the liquid application position to reduce runoff or lateral infiltration; A small wind shield / thin film shed can be arranged above the liquid application position on the soil surface in S1 to stabilize the infiltration boundary and prevent the influence caused by low temperature or dew sensitive conditions.

[0012] Further, the transparent observation module in S1 has several and adopts high light transmission material of 6-10 mm in thickness, the bottom is wedge structure of 10°-20°, the observation window is formed by vertical insertion into soil profile with minimum disturbance, the horizontal spacing is 50-150 mm, and the insertion depth is 250-300 mm; the outer surface can be optionally equipped with anti-condensation heating film or anti-fog / hydrophobic coating; The resolution of the industrial camera in S1 is not less than 1920*1080, and 8-16 mm lens is arranged, fixed exposure, white balance and focusing are arranged to improve the comparability of images; The LED array light source in S1 adopts ring-shaped uniform arrangement and constant current driving, and is equipped with a diffusion cover to improve the illumination uniformity.

[0013] Further, the S4 is supplied with liquid under the condition of constant flow of 10 mL / min; or a step flow scheme of 5→10→15 mL / min is adopted, and the single-stage duration is not less than 5 min.

[0014] Further, in S5, the industrial camera is triggered by the central control and data processing unit to collect images at an interval of 30-60 s / frame, and the LED array light source is controlled to stabilize the illumination; the exposure and white balance of the industrial camera are kept fixed.

[0015] Further, in S6, the collected images are corrected for brightness and color consistency under uniform imaging conditions, and the region of interest is determined; the dyed region is automatically identified and extracted, combined with simple denoising processing, to obtain a stable target region contour; the preferential flow area ratio, preferential flow penetration depth and preferential flow path width are obtained in the target region; a parameter summary table is generated, and the corresponding image is saved.

[0016] The application has the advantages that: 1) The present application is aimed at indoor controlled conditions, and around the "low-temperature constant current liquid supply-in-situ visual imaging-synchronous control and quantitative identification" three links to realize collaborative improvement: in the range of-5℃~+5℃, the tracer solution with freezing point below 0℃ is stably supplied, the surface layer of soil sample is continuously kept wet and the infiltration process is uninterrupted, and the boundary instability caused by near-surface freezing is avoided; through the fixed geometry and exposure combination of wedge-shaped or blade-shaped transparent observation module, industrial camera and ring-shaped illumination, an undamaged in-situ "observation window" is constructed to realize continuous recording of preferential flow initiation, expansion and convergence; the central control and data processing module implements time stamp alignment and closed-loop coordination of pump speed, temperature control, illumination and imaging, and automatically extracts the image results into reusable parameters such as preferential flow area ratio, infiltration depth, path width distribution and fractal characteristics, which is convenient for generating evolution curves over time and cross-sample comparison; compared with the scheme of scattered construction of liquid supply, cooling, imaging and data processing, the present system reduces interface and alignment error, reduces energy consumption and layout complexity, improves the repeatability, comparability and efficiency of the test, and provides a standardized data basis for mechanism research, parameter inversion and model calibration under indoor bench / transparency soil column conditions. BRIEF DESCRIPTION OF DRAWINGS

[0017] The present application will be further described in detail below in combination with the drawings and specific embodiments.

[0018] Figure 1 A flowchart of a low-freezing-point infiltration and visualization integrated detection method for improved saline soil in the seasonal frozen region according to the present application.

[0019] Figure 2 A structure connection relationship diagram of a low-freezing-point infiltration and visualization integrated detection method for improved saline soil in the seasonal frozen region according to the present application.

[0020] Figure 3 A liquid circuit principle diagram of a low-temperature constant current liquid supply module of a low-freezing-point infiltration and visualization integrated detection method for improved saline soil in the seasonal frozen region according to the present application.

[0021] Figure 4 A structure diagram of an in-situ image acquisition module of a low-freezing-point infiltration and visualization integrated detection method for improved saline soil in the seasonal frozen region according to the present application.

[0022] Figure 5 A structure diagram of a central control and data processing module of a low-freezing-point infiltration and visualization integrated detection method for improved saline soil in the seasonal frozen region according to the present application.

[0023] Figure 6 A preferential flow dyeing pixel comparison chart.

[0024] Figure 7 A preferential flow infiltration depth comparison chart.

[0025] Figure 8 For the priority flow path width contrast chart.

[0026] Figure 9 For the transparent observation module side profile binary mask (experimental group). Black is the identified priority flow dyeing area, superimposed with scale calibration and depth reference, used to calculate area ratio and penetration depth.

[0027] Figure 10 For the transparent observation module side profile binary mask (control group). Black is the identified priority flow dyeing area, superimposed with scale calibration and depth reference, used to calculate area ratio and penetration depth. DETAILED DESCRIPTION

[0028] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0029] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present application.

[0030] As Figures 1 to 10 shown in a low freezing point infiltration and visual integrated detection method for improving saline soil in the seasonal frozen region, the specific detection method is as follows: S1: module preparation: Set up a low-temperature constant-flow liquid supply module 1 for supplying a tracer solution with a freezing point lower than 0℃ at a constant temperature and constant flow rate and sending the tracer solution to the surface layer of the soil sample; the low-temperature constant-flow liquid supply module 1 includes a refrigeration insulation box 11, a liquid storage tank 12, a peristaltic pump 13, a flow meter 14 and a liquid application assembly 15; the output end of the liquid storage tank 12 is connected to the peristaltic pump 13, the flow meter 14 and the liquid application assembly 15 in sequence through an insulation hose, and the output end of the liquid application assembly 15 outputs the tracer solution into the soil for natural infiltration.

[0031] Set up an in-situ image acquisition module 3 for continuous in-situ imaging of the priority flow without damage; the in-situ image acquisition module includes a transparent observation module 31, an industrial camera 32 and an LED array light source 33.

[0032] The central control and data processing module 4 is connected with the low-temperature constant-current liquid supply module 1 and the in-situ image acquisition module 3 respectively, used for coordinating the timing of liquid supply, illumination and imaging, adjusting the stability of liquid supply based on flow measurement, and processing the collected images to obtain preferential flow characterization parameters.

[0033] S2: Pre-cooling and constant temperature: start the refrigeration insulation box to the set temperature, and pre-cool the tracer liquid in the liquid storage tank to the target temperature; S3: Flow setting and checking: set the target flow of the peristaltic pump by the central control and data processing module, and check and record according to the flow meter reading; S4: Liquid application and natural infiltration: introduce the insulation hose to the liquid application position on the ground, start the liquid supply, and the tracer liquid forms a stable thin layer and naturally infiltrates; S5: Synchronous imaging and illumination: control the industrial camera and LED array light source to work synchronously at the set interval by the central control and data processing module, acquire continuous images at the transparent observation module, and the industrial camera images the cross section of the transparent observation module at a fixed interval to form a time series image; the binarization results are shown in Figure 9 (experimental group) and Figure 10 (control group).

[0034] S6: Data processing and quantitative analysis: perform denoising, brightness normalization and threshold segmentation processing in the central control and data processing module to obtain the binarization mask (see Figure 9 , Figure 10 ), the black area represents the identified preferential flow dyeing area, and the area ratio, maximum / average penetration depth and dominant path width indicators are output, and the time series curve and report are generated.

[0035] The refrigeration insulation box in S1 adopts a double-layer insulation structure, and a semiconductor refrigeration and temperature sensor is built-in to stably control the temperature in the box at a set low temperature of -2℃±0.5℃, ensuring that the tracer liquid does not freeze at low temperature.

[0036] The tracer solution in the liquid storage tank in S1 is an ethylene glycol / propylene glycol-water system, and visual dye is added; the liquid storage tank has the requirements of low-temperature resistance and corrosion resistance, and is provided with a liquid level mark and a quick connection interface.

[0037] The peristaltic pump in S1 provides adjustable constant flow; the pump pipe is made of low-temperature resistant material to adapt to the tracer liquid.

[0038] The flow meter in S1 is electromagnetic or micro-turbine type, used for monitoring instantaneous flow and cumulative amount and recording.

[0039] Multiple liquid application positions are provided on the soil surface in S1, and the liquid application assembly can be optionally provided with a liquid application diffusion head / drip irrigation head to improve the uniformity of liquid application, and 3-5mm thin layer of fine sand can be laid around the liquid application position to reduce runoff or lateral seepage; In S1, a small wind shield / film shed can be set above the liquid supply position on the soil surface to stabilize the infiltration boundary and prevent the impact of low temperature or dew sensitive conditions.

[0040] In S1, the transparent observation module has several sub-millimeter or tempered glass with a thickness of 6-10 mm of high light transmission material, and the bottom is a wedge structure of 10°-20°, which is vertically inserted into the soil profile to form an observation window with minimal disturbance, with a horizontal spacing of 50-150 mm and an insertion depth of 250-300 mm. The outer surface can be equipped with anti-condensation heating film or anti-fog / hydrophobic coating; In S1, the industrial camera has a resolution of not less than 1920x1080, with an 8-16 mm lens, fixed exposure, white balance and focus to improve image comparability. In S1, the LED array light source is arranged in a ring shape and driven by constant current, and a diffuser cover is provided to improve the uniformity of illumination.

[0041] In S4, the liquid is supplied under a constant flow of 10 mL / min; or a step flow scheme of 5→10→15 mL / min is used, and the single stage duration is not less than 5 min.

[0042] In S5, the industrial camera is triggered by the central control and data processing unit to collect images at an interval of 30-60 s / frame, and the LED array light source is controlled to stabilize the illumination; the exposure and white balance of the industrial camera are fixed.

[0043] In S6, the collected images are corrected for brightness and color consistency under uniform imaging conditions, and the region of interest is determined; the dyeing area is automatically identified and extracted, combined with simple denoising processing, to obtain a stable target area profile; the preferential flow area ratio, preferential flow penetration depth, and preferential flow path width are obtained in the target area; the parameter change curve and summary table over time are generated, and the corresponding images are saved.

[0044] Implementation effect:

[0045] 1) Stability of liquid supply under low temperature conditions. Under indoor controlled conditions, when the temperature in the box is set to -2 ℃ (±0.5 ℃) and the constant flow is 10 mL / min for continuous operation for 30 min, the system realizes continuous liquid supply, and the flow variation coefficient (CV) is not more than 5%, without interruption caused by freezing.

[0046] (2) Recognizability of in-situ imaging. Under the fixed imaging conditions of the transparent observation module and coaxial illumination, the obtained images can identify the preferential flow front and main path, and the spatial resolution can meet the needs of about 1.0 mm / pixel order of morphological identification.

[0047] (3) The repeatability of parameter output. Under the same sample and the same working condition, the area ratio, the maximum / average penetration depth and the path width calculated by the system have good consistency, and the error is in the order of millimeter and individual percentage.

[0048] (4) The operability of the test process. The liquid supply, lighting and imaging are uniformly coordinated by the central control and data processing unit, and the image and process data are recorded synchronously, supporting long-term operation and batch testing at a set interval.

[0049] (5) Applicability. The system is suitable for indoor bench and transparent soil column scenes, and can carry out low-temperature infiltration and preferential flow visualization test in the range of -5 ℃ to +5 ℃, and output quantitative parameters for mechanism research and scheme comparison.

[0050] Under the conditions of setting the temperature in the box to -2 ℃ (±0.5 ℃) and constant flow of 10 mL / min, low-ice-point infiltration and preferential flow visualization tests were carried out on the improved saline soil in the experimental group and the control group in the seasonal frozen region. The image recognition results of a typical test are shown in Table 1 and Figs. 6-8. The total number of pixels in the dyeing area in the two groups is C_tot = 323, but the fully dyed pixels in the experimental group C_full = 135, and in the control group 77; the half-dyed pixels C_half are 12 and 13, respectively, and the quarter-dyed pixels C_quarter are 8 and 10, respectively. The preferential flow dyeing area ratio is calculated as AreaRatio = 0.4427 in the experimental group and AreaRatio = 0.2662 in the control group, and the preferential flow dyeing area ratio in the experimental group is significantly larger than that in the control group, indicating that under the same working conditions, the improvement measures significantly improve the visible area of the profile occupied by the preferential flow (see Figure 6 ).

[0051] Under the same working conditions, 5 representative preferential flow paths were selected, and their penetration depths d1-d5 were 94 mm, 92 mm, 95 mm, 94 mm and 93 mm in the experimental group, and 60 mm, 55 mm, 57 mm, 51 mm and 54 mm in the control group. The corresponding average penetration depth Davg is 93.6 mm in the experimental group and 55.4 mm in the control group, and the maximum penetration depth Dmax is 95 mm and 60 mm, respectively. The results show that the preferential flow front in the experimental group can stably advance to about 0.10 m in depth, while in the control group it is only about 0.06 m, and the comparison of the penetration depths is shown in Fig. 7.

[0052] The corresponding path width W1-W5 in the experimental group is 2.3 mm, 2.5 mm, 2.5 mm, 2.7 mm and 2.4 mm respectively, and the control group is 2.0 mm, 2.1 mm, 2.3 mm, 2.2 mm and 2.0 mm respectively; the average width W_mean of the experimental group is 2.48 mm, and the median width W_median is 2.5 mm, and the control group is 2.12 mm and 2.1 mm respectively. It can be seen that the dominant path width of the experimental group is concentrated in the order of 2-3 mm, and the overall width is wider than that of the control group, and the width contrast is shown in Figure 8.

[0053] The above comparison results show that the low-ice-point infiltration and visual integrated detection system constructed by the application can realize continuous and stable infiltration under the condition of -2 ℃, and significantly improve the preferential flow dyeing area ratio, penetration depth and path width compared with the control group. The geometric characteristics of preferential flow (area ratio, penetration depth and path width, etc.) are clearly and quantitatively characterized, which provides reliable data support for the mechanism research, parameter calibration and comparison of different improvement schemes of the low-temperature infiltration process of saline soil.

[0054] Table 1: Statistics of preferential flow geometric characteristics

[0055] Those skilled in the art should understand that the application is not limited by the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the application. Without departing from the spirit and scope of the application, various changes and improvements can be made to the application, and these changes and improvements all fall within the scope of the claimed application.

Claims

1. A low freezing point infiltration and visualization integrated detection method for improving saline soil in a seasonal frozen region, characterized in that: The specific detection method is as follows: S1: component module preparation: Set up a low-temperature constant-flow liquid supply module for supplying trace solution with a freezing point below 0℃ at a constant temperature and flow rate and sending the trace solution to the surface layer of the soil sample; the low-temperature constant-flow liquid supply module includes a refrigeration insulation box, a liquid storage tank, a peristaltic pump, a flow meter and a liquid application assembly; the output end of the liquid storage tank is connected to the peristaltic pump, the flow meter and the liquid application assembly in sequence through an insulation hose, and the output end of the liquid application assembly outputs the trace solution into the soil for natural infiltration; Set up an in-situ image acquisition module for non-destructive in-situ continuous imaging of preferential flow; the in-situ image acquisition module includes a transparent observation module, an industrial camera and an LED array light source; The central control and data processing module is connected with the low-temperature constant-flow liquid supply module and the in-situ image acquisition module respectively, for coordinating the timing of liquid supply, lighting and imaging, adjusting the stability of liquid supply based on flow measurement, and processing the collected images to obtain the preferential flow characterization parameters; S2: precooling and constant temperature: start the refrigeration insulation box to the set temperature, and precool the trace liquid in the liquid storage tank to the target temperature; S3: flow rate setting and checking: set the target flow rate of the peristaltic pump by the central control and data processing module, and check and record according to the flow meter reading; S4: liquid application and natural infiltration: introduce the insulation hose to the ground liquid application position, start the liquid supply, and the trace liquid forms a stable thin layer and naturally infiltrates; S5: synchronous imaging and lighting: control the industrial camera and LED array light source to work synchronously at the set interval by the central control and data processing module, and acquire continuous images at the transparent observation module; S6: data processing and quantitative analysis: perform denoising, brightness normalization and threshold segmentation processing in the central control and data processing module, output the area ratio, maximum / average penetration depth and dominant path width indicators, generate time series curves and reports.

2. The low freezing point infiltration and visualization integrated detection method for improving saline soil in the seasonal frozen region according to claim 1, characterized in that: The refrigeration insulation box in S1 adopts a double-layer insulation structure, with a built-in semiconductor refrigeration and temperature sensor, which stably controls the temperature in the box at a set low temperature of -2℃±0.5℃, ensuring that the trace liquid does not freeze at low temperature.

3. The low freezing point infiltration and visualization integrated detection method for improving saline soil in the seasonal frozen region according to claim 1, characterized in that: The trace solution in the liquid storage tank in S1 is an ethylene glycol / propylene glycol-water system, and visual dye is added; the liquid storage tank has the requirements of low-temperature resistance and corrosion resistance, and is provided with a liquid level mark and a quick connection interface.

4. The low freezing point infiltration and visualization integrated detection method for improving saline soil in the seasonal frozen region according to claim 1, characterized in that: The peristaltic pump in S1 provides adjustable constant flow; the pump tube is made of low-temperature resistant material to adapt to the trace liquid.

5. The low freezing point infiltration and visualization integrated detection method for improving saline soil in the seasonal frozen region according to claim 1, characterized in that: The flow meter in S1 is electromagnetic or micro-turbine type, used for monitoring instantaneous flow and cumulative amount and recording.

6. The low freezing point infiltration and visualization integrated detection method for improving saline soil in the seasonal frozen region according to claim 1, characterized in that: A plurality of liquid application positions are provided on the surface of the soil in S1, and the liquid application assembly can be optionally provided with a liquid diffusion head / drip irrigation head to improve the uniformity of liquid application, and 3-5mm thin layer of fine sand can be laid around the liquid application position to reduce runoff or lateral infiltration; A small wind shield / thin film shed can be provided above the liquid application position on the surface of the soil in S1 to stabilize the infiltration boundary and prevent the influence of low temperature or dew sensitive conditions.

7. The integrated detection method for low-ice-point infiltration and visualization of improved saline soil in the seasonal frozen region according to claim 1, characterized in that: The transparent observation module in S1 has several pieces of high-transmittance material with a thickness of 6-10 mm, such as acrylic or tempered glass, and a wedge-shaped structure with an angle of 10°-20° at the bottom to form an observation window by vertically inserting the soil profile with minimal disturbance, with a horizontal spacing of 50-150 mm and an insertion depth of 250-300 mm; the outer surface can be optionally equipped with an anti-condensation heating film or an anti-fog / hydrophobic coating; The industrial camera in S1 has a resolution of no less than 1920x1080, with an 8-16 mm lens, fixed exposure, white balance, and focus to improve image comparability; The LED array light source in S1 is arranged in a ring shape and driven by constant current, with a diffuser cover to improve lighting uniformity.

8. The low freezing point infiltration and visualization integrated detection method for improving saline soil in the seasonal frozen region according to claim 1, characterized in that: In S4, the liquid is supplied at a constant flow rate of 10 mL / min; or a step flow rate scheme of 5→10→15 mL / min is used, with a single-stage duration of no less than 5 min.

9. The low freezing point infiltration and visualization integrated detection method for improving saline soil in the seasonal frozen region according to claim 1, characterized in that: In S5, the central control and data processing unit triggers the industrial camera to collect images at an interval of 30-60 s / frame and controls the LED array light source to provide stable illumination; the exposure and white balance of the industrial camera remain fixed.

10. The low freezing point infiltration and visualization integrated detection method for improving saline soil in the seasonal frozen region according to claim 1, characterized in that: In S6, the collected images are corrected for brightness and color consistency under uniform imaging conditions, and the region of interest is determined; the stained area is automatically identified and extracted, combined with simple denoising processing, to obtain a stable target area contour; the preferential flow area ratio, preferential flow penetration depth, and preferential flow path width are obtained in the target area; A parameter summary table is generated, and the corresponding images are saved.

Images