Mushroom growth light supplement control method and system based on multispectral LED light source

Abstract

The invention relates to a mushroom growth light supplement control method and system based on a multispectral LED light source. The mushroom growth light supplement control method based on the multispectral LED light source comprises the steps that the multispectral LED light source is arranged in a mushroom growth environment; constructing a mushroom growth model database containing corresponding relations of mushroom varieties, growth stages and spectral parameters; acquiring a real-time growth state image of mushrooms; judging the variety and the current growth stage of the mushrooms; judging whether the current growth stage reaches a preset condition for triggering stage conversion or not, if so, calling a mushroom growth model database, and obtaining an optimal spectrum parameter; calculating the PWM duty ratio of each color LED based on the optimal spectrum parameter; a driving current signal is obtained; and performing driving adjustment on the multispectral LED light source based on the driving current signal. The application can meet the differentiated requirements of mushrooms in different growth stages on red, blue and green spectrums, so that the yield and quality of the mushrooms are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mushroom cultivation, and in particular to a mushroom growth light supplementing control method and system based on a multi-spectrum LED light source. BACKGROUND

[0002] With the increasing demand for simultaneous improvement of yield and quality in mushroom factory cultivation, LED light supplementing technology, as a key measure for precise regulation of light environment, has the advantages of low energy consumption, long service life, and strong light spectrum adjustability, and has gradually been widely recognized by the planting industry and has been applied in large-scale cultivation of mainstream mushrooms such as golden needle mushrooms and shiitake mushrooms. However, existing LED light supplementing systems mostly use fixed spectrum design, which cannot dynamically and accurately adjust the spectrum according to the specific light requirements of different mushroom varieties and their different growth stages. SUMMARY

[0003] Therefore, it is necessary to solve the above problems, and the present application provides a mushroom growth light supplementing control method and system based on a multi-spectrum LED light source.

[0004] In a first aspect, the present application provides a mushroom growth light supplementing control method based on a multi-spectrum LED light source, which comprises:

[0005] A multi-spectrum LED light source is provided in a mushroom growth environment, and the multi-spectrum LED light source comprises multi-color LEDs;

[0006] A mushroom growth model database containing the correspondence between mushroom varieties, growth stages, and spectrum parameters is constructed;

[0007] An image of the real-time growth state of the mushroom is obtained;

[0008] The real-time growth state image is preprocessed and feature extracted to determine the variety and current growth stage of the mushroom;

[0009] It is determined whether the current growth stage meets the preset condition for triggering stage conversion, and if so, the mushroom growth model database is called, and the optimal spectrum parameter is obtained based on the mushroom growth model database, the variety of the mushroom, and the current growth stage;

[0010] Based on the optimal spectrum parameter, the PWM duty cycle of each color LED is calculated;

[0011] Based on the PWM duty cycle of each color LED, a driving current signal is obtained;

[0012] The multi-spectrum LED light source is driven and adjusted based on the driving current signal.

[0013] In some embodiments, after obtaining the optimal spectral parameters, before calculating the PWM duty cycles of the LEDs of each color, the method further comprises:

[0014] constructing a spectral superposition equation to calculate the mixing coefficients of the LEDs of each color, wherein the constructed spectral superposition equation comprises:

[0015]

[0016] wherein X, Y, and Z represent the CIE 1931 standard color values; 、 、 are the CIE standard observer color matching functions; 、 、 are the spectral power distribution functions of the LEDs; 、 、 are the mixing coefficients of the red LEDs, blue LEDs, and green LEDs, respectively; is the wavelength of light.

[0017] In some embodiments, the real-time growth state image is preprocessed and feature extracted to determine the variety and current growth stage of the mushroom, comprising:

[0018] the contrast of the real-time growth state image is enhanced using adaptive histogram equalization, and the edge noise of the real-time growth state image is removed to obtain a smooth grayscale image;

[0019] the grayscale image is converted into a binary image containing mycelium and background using an Otsu dynamic threshold segmentation algorithm;

[0020] the binary image is feature extracted to obtain the variety and current growth stage of the mushroom.

[0021] In some embodiments, the feature extraction of the binary image comprises:

[0022] the binary image can be feature extracted by a ResNet-18 network;

[0023] the output end of the ResNet-18 network is divided into three layers, which respectively output the mycelium density, primordium number, and basidiome diameter.

[0024] In some embodiments, the mushroom growth model database is stored in a structured manner, and each record in the mushroom growth model database contains a spectral stage identifier, a variety of mushroom, a growth stage of the mushroom, and corresponding optimal spectral parameters;

[0025] The spectral parameters include a red light proportion coefficient, a blue light proportion coefficient, a green light proportion coefficient, and a target total light intensity.

[0026] In some embodiments, a three-layer control architecture is constructed, including an upper master control chip, a middle driving chip, and the multi-spectral LED light source at the bottom layer.

[0027] The upper master control chip is configured to determine whether the current growth stage reaches a preset condition triggering a stage conversion, and if so, call the mushroom growth model database, obtain optimal spectral parameters based on the mushroom growth model database, the variety of the mushroom, and the current growth stage, and calculate the PWM duty cycle of each color LED based on the optimal spectral parameters.

[0028] The middle driving chip is configured to obtain a driving current signal based on the PWM duty cycle of each color LED, and drive and adjust the multi-spectral LED light source based on the driving current signal.

[0029] In some embodiments, the multi-spectral LED light source further includes:

[0030] A heat dissipation substrate;

[0031] The multi-color LEDs are located on the surface of the heat dissipation substrate and form a red, blue, and green three-primary color LED array. Each color LED array includes a plurality of driving units, and each driving unit includes a plurality of LEDs in parallel.

[0032] A plurality of optical lenses are arranged one-to-one corresponding to the driving units and located on the light output surface of each driving unit.

[0033] In some embodiments, the PWM duty cycle D is obtained based on the following formula:

[0034] ;

[0035] Wherein, D is the PWM duty cycle; is the proportion coefficient of the corresponding color LED; is the target total light intensity; is the maximum light intensity of the corresponding color LED; is the light efficiency decay coefficient of the corresponding color LED; is the 16-bit PWM maximum resolution.

[0036] In some embodiments, after the multi-spectral LED light source is driven and adjusted based on the driving current signal, the method further includes repeating the following steps several times:

[0037] Real-time monitoring of the actual output light signal of each color LED;

[0038] correct the optimal spectrum parameter based on the spectrum superposition equation if the optical signal is detected to be abnormal;

[0039] calculate an updated PWM duty cycle based on the corrected optimal spectrum parameter;

[0040] update the driving current signal based on the updated PWM duty cycle;

[0041] drive and adjust the multi-spectrum LED light source based on the updated driving current signal.

[0042] In a second aspect, the present application also provides a mushroom growth light supplement control system based on a multi-spectrum LED light source, which comprises:

[0043] a spectrum LED light source arranged in a mushroom growth environment, the multi-spectrum LED light source comprising multi-color LEDs;

[0044] a database construction module for constructing a mushroom growth model database comprising a correspondence relationship among mushroom varieties, growth stages, and spectrum parameters;

[0045] a collection module for acquiring real-time growth state images of mushrooms;

[0046] a growth state recognition module for pre-processing and feature extraction of the real-time growth state images to determine mushroom varieties and current growth stages;

[0047] a spectrum parameter calling module for determining whether the current growth stage meets a preset condition for triggering stage conversion, and if so, calling the mushroom growth model database and obtaining optimal spectrum parameters based on the mushroom growth model database, the variety of the mushroom, and the current growth stage;

[0048] a multi-spectrum LED regulation module for calculating PWM duty cycles of each color LED based on the optimal spectrum parameters, obtaining driving current signals based on the PWM duty cycles of each color LED, and driving and adjusting the multi-spectrum LED light source based on the driving current signals.

[0049] The mushroom growth light supplement control method and system based on a multi-spectrum LED light source according to the present application can automatically match optimal spectrum parameter combinations according to mushroom varieties and growth stages, can ensure the stability and uniformity of light in a mushroom growth environment, and can meet the differentiated needs for red, blue, and green spectrums at different growth stages through precise proportioning control of multi-spectrum LEDs, thereby significantly improving the yield and quality of mushrooms. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0051] Figure 1 The flow chart of the method for controlling light supplement for mushroom growth based on a multi-spectrum LED light source provided in an embodiment of the present application;

[0052] Figure 2 The structural block diagram of the system for controlling light supplement for mushroom growth based on a multi-spectrum LED light source provided in another embodiment of the present application.

[0053] Legend: 10, spectrum LED light source; 20, database construction module; 30, acquisition module; 40, growth state recognition module; 50, spectrum parameter calling module; 60, multi-spectrum LED regulation and control module. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of the present application more clear, the present application will be further described in detail below in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0055] In one embodiment, referring to Figure 1 The present application provides a method for controlling light supplement for mushroom growth based on a multi-spectrum LED light source, which comprises the following steps: S11-S18.

[0056] S11: A multi-spectrum LED light source is provided in a mushroom growth environment, wherein the multi-spectrum LED light source comprises multi-color LEDs.

[0057] S12: A mushroom growth model database containing the corresponding relationship among mushroom varieties, growth stages and spectrum parameters is constructed.

[0058] S13: Real-time growth state images of mushrooms are acquired.

[0059] S14: The real-time growth state images are pre-processed and feature-extracted to determine the variety and current growth stage of the mushrooms.

[0060] S15: It is determined whether the current growth stage meets the preset condition for triggering stage conversion. If yes, the mushroom growth model database is called, and the optimal spectrum parameters are obtained based on the mushroom growth model database, the variety and current growth stage of the mushrooms.

[0061] S16: Calculate the PWM duty cycle of each color LED based on the optimal spectral parameter.

[0062] S17: Obtain a driving current signal based on the PWM duty cycle of each color LED.

[0063] S18: Drive and adjust the multi-spectral LED light source based on the driving current signal.

[0064] The mushroom growth light supplement control method based on the multi-spectral LED light source of the present application can automatically match the optimal spectral parameter combination according to the mushroom variety and growth stage; can ensure the stability and uniformity of the light in the mushroom growth environment; through the precise ratio control of the multi-spectral LED, the differentiated demand for red, blue and green spectrum at different growth stages can be met, thereby significantly improving the yield and quality of the mushrooms. It should be noted that the growth environment of the mushrooms in the present application is a growth environment without natural light, and the light required for the growth of the mushrooms is provided by the multi-spectral LED light source.

[0065] In step S11, please refer to Figure 1 of the S11 step, a multi-spectral LED light source is arranged in the mushroom growth environment, and the multi-spectral LED light source includes multi-color LEDs.

[0066] As an example, the multi-spectral LED light source can also include:

[0067] a heat dissipation substrate (not shown);

[0068] The multi-color LEDs (not shown) are all located on the surface of the heat dissipation substrate and form a red, blue and green three-primary color LED array; each color LED array includes a plurality of driving units; each driving unit includes a plurality of parallel LEDs;

[0069] A plurality of optical lenses (not shown) are arranged one-to-one corresponding to the driving units and are located on the light emitting surface of each driving unit.

[0070] Specifically, the heat dissipation substrate is the physical heat dissipation base of the LED array, which can be sized according to actual needs, for example, a 6061 aluminum alloy substrate (thickness 2mm) can be cut according to the size of 350mm×350mm, and the edge chamfer (R2mm) is avoided to avoid scratching; then the upper surface of the substrate can be anodized to form an insulating layer of 50μm thick to ensure the safety of the circuit; and the lower surface is polished to Ra≤0.8μm, 1mm thick thermal conductive silicone grease (thermal conductivity≥3.0W / (m·K)) is applied, and the bonding surface with the external heat dissipation LED array is reserved. The front surface of the heat dissipation substrate can be used to arrange the multi-color LEDs, and the back surface of the heat dissipation substrate can be provided with heat dissipation fins for heat dissipation to improve the heat dissipation efficiency.

[0071] Specifically, the circuit layout can be designed as a 32-channel driving channel, and 4 LEDs of the same color can be grouped as a driving unit, and a common anode connection method is adopted. Each driving unit is provided with a test point for subsequent measurement of current size. Then, a hole (Φ4mm) is drilled at each corner of the driving unit on the aluminum substrate, and a positioning column (Φ2mm) is fixed and installed for accurate positioning of the lens. Then, the optical lens (for example, a PMMA lens with a light-shielding skirt) is buckled on the positioning column, and the distance between the bottom surface of the optical lens and the light-emitting surface of the LED is 3-5mm (optimized according to the beam angle: when the distance is 4mm, the beam angle is exactly 120°). During the adjustment of the height of the optical lens, the light intensity distribution instrument can be used for detection to ensure that the light of each driving unit forms a light spot with a diameter of 2.1m at a distance of 1m (corresponding to a 120° beam angle), and the light intensity difference between the center and the edge of the light spot is less than 5%. In addition, black silicone (temperature resistance ≥100℃) can be filled between the light-shielding skirts of the optical lens to form a light-shielding isolation belt, thereby avoiding light crosstalk between adjacent driving units (crosstalk rate <2%).

[0072] Specifically, a red (620-660nm), blue (450-470nm), and green (520-530nm) three-primary-color LED array is used as the core light source, and 4 LEDs of the same color are grouped as a driving unit. A common anode parallel connection method (anodes are connected together, and cathodes are connected to the driving channel through current-limiting resistors, respectively) is adopted to ensure uniform distribution of the current of each LED (for example, 60mA channel current corresponds to 15mA per LED). Then, the driving units are welded on the aluminum substrate circuit layout in the order of red, blue, and green alternately to form an 8x8 matrix, and the center-to-center distance between adjacent driving units is 50mm. During the welding process, the LED bottom heat dissipation pad needs to be tightly attached to the aluminum substrate (for example, a reflow soldering process is adopted, and the temperature is 260℃±5℃) to ensure that the heat is directly conducted to the substrate. After welding, the positive and negative resistances of each driving unit can be measured with a multimeter, and the single-channel LED light-emitting consistency can also be tested.

[0073] In step S12, please refer to the S12 step in Figure 1 , a mushroom growth model database containing the correspondence relationship between mushroom varieties, growth stages, and spectral parameters is constructed.

[0074] As an example, the mushroom growth model database is stored in a structured manner, and each record in the mushroom growth model database contains a spectral stage identifier (for example, a spectral stage ID), a mushroom variety, a mushroom growth stage, and corresponding optimal spectral parameters.

[0075] As an example, the spectral parameters can include a red light ratio coefficient , a blue light ratio coefficient Green light ratio coefficient and total luminous intensity of the target (Unit: μmol / m²·s). Wherein, the red light sizing coefficient... Blue light ratio coefficient Green light ratio coefficient The value range of can be (0,1).

[0076] Specifically, the core elements of the database can be defined by combining the needs of industrialized mushroom production with the characteristics of the system hardware. In terms of varieties, priority should be given to commercially mainstream enoki and shiitake mushrooms, with space reserved for expansion to include king oyster mushrooms, oyster mushrooms, etc. Based on the biological characteristics of mushrooms, the growth cycle should be divided into quantifiable stages (e.g., enoki mushrooms: mycelial germination stage → mycelial spread stage → primordium formation stage → fruiting body elongation stage → maturity stage; shiitake mushrooms: mycelial stage → primordium differentiation stage → bud stage → fruiting body enlargement stage), with each stage defined by clear morphological characteristics (e.g., mycelial density ≥80%, primordium diameter ≥2mm). The core parameters that need to be stored should be clearly defined, mainly including the red light ratio coefficient. Blue light ratio coefficient Green light ratio coefficient and total luminous intensity of the target .

[0077] Specifically, a constant temperature and humidity incubation chamber can be built (e.g., temperature 20±1℃, humidity 85±5%). The concentration is 500±50ppm (or a conventional mushroom cultivation chamber, but this chamber must be shaded to prevent interference from natural light) to eliminate interference from other environmental factors on growth; and a multispectral LED light source consistent with the system hardware (red light 620-660nm, blue light 450-470nm, green light 520-530nm, 8×8 matrix layout) is used to ensure that the experimental spectrum is consistent with the actual system output spectrum. For each growth stage, gradient combinations of red light ratio coefficients, blue light ratio coefficients, and green light ratio coefficients are set (for example, each parameter is taken at 0.1 intervals, for a total of 10×10×10=1000 combinations, which are reduced to 50-80 key combinations through orthogonal experimental methods). Each combination can be repeated 3 times to avoid random errors. Subsequently, consistent image recognition technology (5-megapixel camera + 850nm supplemental lighting) was used to collect morphological indicators such as mycelial density, primordium quantity, and fruiting body diameter daily; and physiological indicators (such as mycelial dry weight growth rate and fruiting body vitamin D content) were monitored simultaneously to comprehensively evaluate the impact of spectral parameters on mushroom growth rate and quality.

[0078] Specifically, quantitative analysis can be performed according to historical experimental data to screen out the optimal spectrum parameters for each variety-stage. Different indicators (such as mycelium density 0-1, fruiting body diameter 0-50mm) can be normalized to the interval [0,1] to eliminate dimensional differences; according to the planting target (such as commodity mushroom focusing on growth rate and fruiting body size), weights are assigned to each indicator (for example, mycelium stage: growth rate weight 0.8, others 0.2; mature stage: quality indicator weight 0.6); then the comprehensive score of each group of spectrum parameters is calculated (by weighted sum of each indicator normalized value), 1-3 groups of parameters with the highest comprehensive score are selected, and their significance (P<0.05) is verified by analysis of variance (ANOVA), and finally the only optimal spectrum parameter is determined.

[0079] Specifically, the screened optimal parameters are stored in the database according to the principle of quick calling. The database uses structured storage, and each record contains spectrum stage identification (such as 101-Enoki mushroom-mycelium stage), mushroom variety (such as Enoki mushroom), growth stage (such as mycelium stage), and target spectrum parameters. The spectrum parameters can include red light ratio coefficient , blue light ratio coefficient , green light ratio coefficient , and target total light intensity . In addition, the database needs to be continuously optimized through actual planting feedback to ensure the applicability of the parameters; in the large-scale planting scene, the light supplement is controlled according to the database parameters, and the deviation between the actual growth indicators and the experimental data is corrected to update the parameters in the database.

[0080] As an example, the mushroom growth model database established in step S12 through experimental demonstration, quantitative analysis, field verification, and dynamic iteration ensures that the corresponding relationship between the stored mushroom variety-growth stage-spectrum parameters has scientific basis and practical application value.

[0081] In step S13, please refer to the S13 step in Figure 1 to obtain the real-time growth state image of the mushroom.

[0082] As an example, a high-resolution industrial camera (at least 5 million pixels) is used with an 850nm near-infrared fill light to eliminate visible light interference; the high-resolution industrial camera can be set to a collection frequency of 30fps, and a timing and on-demand double trigger mode is used, that is, 1 frame of image is collected every 20 seconds under normal conditions (balance real-time and computing power consumption); when the system detects that the spectrum parameter is about to switch, 10 consecutive frames of collection are triggered, and the middle frame is taken to avoid accidental interference, and the real-time growth state image of the mushroom is obtained.

[0083] In step S14, please refer to Figure 1S14: Preprocessing and feature extraction are performed on the real-time growth state image to determine the variety and current growth stage of the mushroom.

[0084] As an example, preprocessing and feature extraction are performed on the real-time growth state image to determine the variety and current growth stage of the mushroom, which can include the following steps:

[0085] S141: Adaptive histogram equalization is used to enhance the contrast of the real-time growth state image; edge noise of the real-time growth state image is removed to obtain a smooth gray image.

[0086] S142: The Otsu dynamic threshold segmentation algorithm is used to convert the gray image into a binary image containing mycelium and background.

[0087] S143: Feature extraction is performed on the binary image to obtain the variety and current growth stage of the mushroom.

[0088] Specifically, adaptive histogram equalization (CLAHE) can be used to enhance the contrast, and a limited contrast histogram equalization is applied to each 8x8 pixel block to enhance the edge features of the mycelium and substrate; the contrast limit parameter (clipLimit) is set to 2.0 to limit the contrast gain of the local histogram, avoid noise amplification, and enhance the gray difference between the mushroom and the culture medium; the block size (tileGridSize) is set to (8,8), i.e. the image is divided into 8x8 tiles, and histogram equalization is performed on each tile independently to adapt to the local brightness difference caused by uneven illumination in the culture space and improve the clarity of edge details such as the outline of the fruiting body.

[0089] Specifically, for the sensor dark current noise and light compensation artifacts of the image edge, a 3x3 median filter algorithm can be used for processing. Taking each pixel as the center, the gray value of the 3x3 neighborhood is replaced by the median value, effectively removing the isolated noise points of the image, and preserving the original base outline and other fine edges, avoiding excessive filtering that leads to feature loss; the edge noise of the processed image is reduced by more than 90%, the gray transition is continuous, and there are no obvious artifacts, forming a smooth gray image.

[0090] Specifically, the optimal threshold can be calculated based on the maximum inter-class variance method (Otsu) to binarize the image. First, the gray histogram is counted, and all threshold values t from 0 to 255 are traversed, and the intra-class variance can be calculated based on the following formula, including:

[0091]

[0092] where t is the segmentation threshold value. a proportion of the total pixels that are pixels of the first type; a proportion of the total pixels that are pixels of the second type; a variance within the first type; a variance within the second type.

[0093] As an example, the optimal threshold can be determined by maximizing the inter-class variance or minimizing the intra-class variance for automatic segmentation of mycelium and culture medium background; then a "3x3 erosion-dilation" morphological operation is performed to remove isolated noise points and fill in small cavities to obtain a binary image with a mycelium and background separation accuracy of more than 95%.

[0094] As an example, feature extraction on the binary image can include the following: the binary image can be feature-extracted by a ResNet-18 network; the output end of the ResNet-18 network is divided into three layers, which respectively output mycelium density (0-1), primordium number (pieces / cm 2 ), and basidiome diameter (mm).

[0095] Specifically, a spatial attention module (SAM) is added after the third residual block to strengthen the extraction of mycelium distribution features, and the calculation expression of the spatial attention module can be:

[0096]

[0097] wherein, is a spatial attention weight map; F is an input feature image; is max pooling; is average pooling; is compressed and fused features; is an activation function.

[0098] Specifically, the output layer is set to a three-head structure, which respectively predicts mycelium density (0-1), primordium number (pieces / cm²), and basidiome diameter (mm); in addition, a focal loss function (Focal Loss) is used during training to solve the class imbalance problem, and the expression of the focal loss function can be:

[0099]

[0100] wherein, is a model predicted sample; is a balance factor; is a difficult sample focus parameter.

[0101] Specifically, based on the quantification indicators output by the ResNet-18 network, the threshold of the mushroom growth model database is combined for judgment; the growth characteristics of different mushroom varieties have inherent differences (for example, the fruiting body of golden needle mushroom is slender, and the fruiting body of shiitake mushroom is thick and strong), which can be matched with the database template through the combination of fruiting body diameter-mycelium density characteristics, for example, fruiting body diameter 5-15 mm and mycelium density > 0.8, which is determined as golden needle mushroom; fruiting body diameter ≥ 20 mm and mycelium density 0.7-0.9, which is determined as shiitake mushroom, and the variety recognition accuracy is ≥ 98%. The judgment of the growth stage can be matched by setting the quantification threshold, for example, the mycelium stage of golden needle mushroom: mycelium density < 0.8, primordium number = 0, and fruiting body diameter = 0; the fruiting body induction stage of golden needle mushroom: mycelium density ≥ 0.8, primordium number 0-5 / cm², and fruiting body diameter < 2 mm; and finally output the variety-growth stage label (for example, golden needle mushroom-fruiting body elongation stage) to trigger the spectrum parameter calling.

[0102] In step S15, please refer to Figure 1 the S15 step in the above formula, determine whether the current growth stage reaches the preset condition of triggering stage conversion, if yes, call the mushroom growth model database, and get the optimal spectrum parameter based on the mushroom growth model database, the variety of the mushroom and the current growth stage.

[0103] Specifically, the mycelium density, primordium number, and fruiting body diameter output by the ResNet-18 network are used as the core dimensions, and the preset quantification threshold is combined for judgment.

[0104] Specifically, the stage conversion judgment conditions of golden needle mushroom are: mycelium stage → fruiting body induction stage: mycelium density ≥ 0.8; primordium number = 0; fruiting body diameter = 0; fruiting body induction stage → fruiting body elongation stage: mycelium density ≥ 0.85; primordium number ≥ 3 / cm 2 ; fruiting body diameter ≥ 2 mm; fruiting body elongation stage → mature stage: mycelium density ≥ 0.9; primordium number ≥ 2 / cm²; fruiting body diameter ≥ 10 mm.

[0105] Specifically, the stage conversion judgment conditions of shiitake mushroom are: mycelium stage → primordium differentiation stage: mycelium density ≥ 0.85; primordium number ≥ 5 / cm²; fruiting body diameter = 0; primordium differentiation stage → bud stage: mycelium density ≥ 0.8; primordium number ≥ 3 / cm²; fruiting body diameter ≥ 2 mm; bud stage → fruiting body swelling stage:

[0106] mycelium density ≥ 0.75; primordium number ≥ 1 / cm²; fruiting body diameter ≥ 5 mm.

[0107] More specifically, taking Flammulina velutipes as an example, if the current stage of Flammulina velutipes is the fruit body induction period, the real-time features are 2.5 primordia per cm2 and 1.8 mm in diameter of fruit body; the conditions of matching the fruit body induction period to the fruit body elongation period are mainly used to determine the mycelium period to the fruit body induction period, to ensure the maturity of the mycelium; the conversion at this stage needs to take the development of primordia as the core, that is, the number of primordia and the diameter of fruit body, to avoid the interference of the stable mycelium density on the accurate determination of the reproductive growth stage; at this time, the number of primordia is 2.5, which is less than 3 (not up to standard), and the diameter of fruit body is 1.8, which is less than 2 (not up to standard); it is determined that the conversion is not triggered, the current stage is maintained, and the original spectral parameters are continued to be used for light supplement; if the current stage of Flammulina velutipes is the mycelium period, the real-time features are 0.82 mycelium density, 0 primordia, and 0 diameter of fruit body; the preset conditions of matching the mycelium period to the fruit body induction period are matched one by one, the mycelium density is 0.82, which is greater than 0.8 (up to standard), the number of primordia is 0=0 (up to standard), and the diameter of fruit body is 0=0 (up to standard); at this time, all conditions are met, the stage conversion is triggered, and the target stage is updated to the fruit body induction period.

[0108] Specifically, after triggering the stage conversion, the Flammulina velutipes growth model database is called based on the variety and target growth stage of the Flammulina velutipes to obtain the optimal spectral parameters of the variety and current growth stage of the Flammulina velutipes.

[0109] In step S16, please refer to the S16 step in Figure 1 Based on the optimal spectral parameters, the PWM duty cycles of each color LED are calculated.

[0110] As an example, the Flammulina velutipes growth light supplement method based on the multi-spectral LED light source can further include the following contents: a three-layer control architecture is constructed, which can include an upper layer master control chip, a middle layer driving chip, and the multi-spectral LED light source at the bottom layer; the upper layer master control chip is used to determine whether the current growth stage meets the preset conditions for triggering the stage conversion, if yes, the Flammulina velutipes growth model database is called, the optimal spectral parameters are obtained based on the Flammulina velutipes growth model database, the variety and current growth stage of the Flammulina velutipes; and the PWM duty cycles of each color LED are calculated based on the optimal spectral parameters; the middle layer driving chip is used to obtain driving current signals based on the PWM duty cycles of each color LED, and to drive and adjust the multi-spectral LED light source based on the driving current signals.

[0111] Specifically, the upper layer master control chip can be an upper layer STM32F407 master control chip, and the middle layer driving chip can be a cascaded TLC5971 driving chip; the upper layer master control chip as the core control unit can establish a communication link with the two cascaded TLC5971 driving chips in the middle layer through an I2C interface at a rate of 400 kHz.

[0112] Specifically, two cascaded TLC5971 driver chips are expanded to 32 constant current outputs through cascading (through SIN / SOUT pins in series); for example, the 1-10 channels of the cascaded TLC5971 driver chip are connected to the red light driving unit, the 11-21 channels are connected to the blue light driving unit, and the 22-32 channels are connected to the green light driving unit; each driving channel is connected in series with a 10Ω±1% precision current sampling resistor for real-time current feedback calibration.

[0113] Specifically, the digital machine interface of the upper master control chip can be directly connected to an OV5640 high-resolution industrial camera (≥5 million pixels), which is deployed 1.5-2m above the mushrooms, and cooperates with an 850nm near-infrared fill light to eliminate environmental light interference to collect mushroom growth state images; in addition, a spectrometer (detection wavelength 400-700nm, accuracy ±3%) is fixed 1.5m below the multi-spectral LED light source, and the data output end is connected to the master control chip to realize output spectrum monitoring and correction.

[0114] Specificly, PWM dimming control program can be configured in STM32F407 master control chip, the dimming frequency in PWM dimming parameters is set to 1kHz to avoid flicker interference, and the duty cycle resolution is 16 bits (0-65535) to ensure 0.1% light intensity adjustment accuracy.

[0115] As an example, after obtaining the optimal spectral parameters, before calculating the PWM duty cycle of each color LED, the following content can also be included:

[0116] The spectral superposition equation is constructed to calculate the matching coefficient of each color LED; the constructed spectral superposition equation includes:

[0117]

[0118] Where X, Y, and Z represent the CIE 1931 standard excitation value; 、 、 is the CIE standard observer color matching function; 、 、 is the spectral power distribution function of each LED; 、 、 are the matching coefficients of red, blue, and green LEDs, respectively; is the wavelength of light.

[0119] As an example, the PWM duty cycle D can be obtained based on the following formula:

[0120] ;

[0121] wherein D is the PWM duty cycle; is the matching coefficient of the corresponding color LED; is the target total light intensity; is the maximum light intensity of the corresponding color LED; is the light efficiency decay coefficient of the corresponding color LED; is the 16-bit PWM maximum resolution.

[0122] Specifically, the matching coefficient of each channel is obtained based on the mushroom growth model database 、 、 and the target total light intensity (for example, the fruiting body stage of the golden needle mushroom =0.4, =0.3, =0.3, =800μmol / m²·s), the maximum light intensity value of each channel is determined from the light intensity-current curve in the LED specification book (for example, red light 1200μmol / m²·s, blue light 800μmol / m²·s, green light 600μmol / m²·s), and the light efficiency decay coefficient is retrieved from the operation log (for example, blue light is used for 2000 hours _B=0.95); then, the PWM duty cycle of each color LED is calculated by substituting the parameters of the red, blue, and green channels into the formula; for example, the fruiting body stage of the golden needle mushroom (new lamp =1.0) calculates D_R=17476, D_B=19661, and D_G=26214; and the primordium differentiation stage of the shiitake mushroom (blue light =0.95) calculates D_R=0, D_B=28876, and D_G=18043.

[0123] In step S17, please refer to the S17 step in Figure 1 , the driving current signal is obtained based on the PWM duty cycle of each color LED.

[0124] Specifically, the STM32F407 master chip transmits the calculated red, blue, and green three-channel PWM duty cycle values to the cascaded TLC5971 driving chip in the middle layer through the I²C interface at a rate of 400kHz; then, the cascaded TLC5971 driving chip receives the digital duty cycle signal, the internal circuit analyzes the duty cycle value and the corresponding channel, and converts it into a precise analog current output driving current signal through the constant current source module.

[0125] In step S18, referring to Figure 1 , the multi-spectrum LED light source is driven and adjusted based on the driving current signal.

[0126] Specifically, the driving current signal acts on the multi-spectrum LED light source, and red, blue, and green spectrums output light intensity according to corresponding proportions, and cooperate with the exclusive optical lens of each channel to realize uniform irradiation. At the same time, the aluminum substrate heat dissipation design controls the junction temperature of the multi-spectrum LED light source below 60°C, avoiding the current-light intensity characteristic drift caused by temperature rise.

[0127] As an example, after the multi-spectrum LED light source is driven and adjusted based on the driving current signal, it can further include repeating the following steps several times: real-time monitoring of the light signal actually output by each color LED; if an abnormal light signal is detected, correcting the optimal spectral parameters based on the spectral superposition equation; calculating and updating the PWM duty cycle based on the corrected optimal spectral parameters; updating the driving current signal based on the updated PWM duty cycle; and driving and adjusting the multi-spectrum LED light source based on the updated driving current signal.

[0128] Specifically, the light signal output by the multi-spectrum LED light source is monitored in real time by a spectrometer, and the spectral composition and actual light intensity of red (620-660nm), blue (450-470nm), and green (520-530nm) are collected. If an abnormal light signal is detected (error ≥±3% from the target value), the STM32F407 master control chip corrects the red, blue, and green channel ratio coefficients based on the spectral superposition equation , , to obtain the corrected optimal spectral parameters. Then the PWM duty cycle of each color LED is calculated and updated according to the formula and updated to the cascaded TLC5971 driving chip to obtain a new driving current signal, and the multi-spectrum LED light source is driven and adjusted based on the driving current signal. At this time, the spectrometer monitors the actual output light signal again, and the above steps are repeated until the light signal error is less than 3%, realizing dynamic stability of the multi-spectrum lighting environment.

[0129] In another embodiment, referring to Figure 2The application also provides a mushroom growth light supplement control system based on a multi-spectrum LED light source. The mushroom growth light supplement control system based on the multi-spectrum LED light source can include a spectrum LED light source 10, a database construction module 20, an acquisition module 30, a growth state recognition module 40, a spectrum parameter calling module 50, and a multi-spectrum LED regulation module 60. The spectrum LED light source 10 is arranged in a mushroom growth environment, and the multi-spectrum LED light source includes multi-color LEDs. The database construction module 20 is configured to construct a mushroom growth model database containing the corresponding relationship among mushroom varieties, growth stages, and spectrum parameters. The acquisition module 30 is configured to obtain real-time growth state images of mushrooms. The growth state recognition module 40 is configured to pre-process and extract features from the real-time growth state images to determine the mushroom variety and the current growth stage. The spectrum parameter calling module 50 is configured to determine whether the current growth stage meets a preset condition for triggering stage conversion. If yes, the mushroom growth model database is called, and the optimal spectrum parameters are obtained based on the mushroom growth model database, the mushroom variety, and the current growth stage. The multi-spectrum LED regulation module 60 is configured to calculate the PWM duty cycle of each color LED based on the optimal spectrum parameters, obtain a driving current signal based on the PWM duty cycle of each color LED, and drive and adjust the multi-spectrum LED light source based on the driving current signal.

[0130] In the mushroom growth light supplement control system based on the multi-spectrum LED light source, the spectrum LED light source 10, the database construction module 20, the acquisition module 30, the growth state recognition module 40, the spectrum parameter calling module 50, and the multi-spectrum LED regulation module 60 can automatically match the optimal spectrum parameter combination according to the mushroom variety and the growth stage, ensure the stability and uniformity of the light in the mushroom growth environment, and meet the differentiated requirements for red, blue, and green spectrum at different growth stages through precise proportioning control of the multi-spectrum LED, thereby significantly improving the yield and quality of mushrooms.

[0131] As an example, the mushroom growth light supplement control system based on the multi-spectrum LED light source can be used to perform the mushroom growth light supplement control method based on the multi-spectrum LED light source as described in the embodiments and related embodiments. Figure 1

[0132] The technical features of the above embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features of the above embodiments are described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present disclosure.

[0133] ​The above embodiments only express several implementation ways of the present application, and the description is specific and detailed, but it should not be understood as a limitation to the patent scope of the application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for controlling supplemental lighting for mushroom growth based on a multispectral LED light source, characterized in that, include: A multispectral LED light source is set in the mushroom growth environment, wherein the multispectral LED light source includes multicolor LEDs; Construct a mushroom growth model database containing the correspondence between mushroom varieties, growth stages, and spectral parameters; Acquire real-time images of the growth status of mushrooms; The real-time growth status images are preprocessed and feature extracted to determine the variety and current growth stage of the mushrooms. Determine whether the current growth stage has reached the preset conditions for triggering stage transition. If so, call the mushroom growth model database and obtain the optimal spectral parameters based on the mushroom growth model database, the mushroom variety, and the current growth stage. Based on the optimal spectral parameters, the PWM duty cycle of each color LED is calculated; The drive current signal is obtained based on the PWM duty cycle of each color LED; The multispectral LED light source is driven and adjusted based on the driving current signal.

2. The method according to claim 1, characterized in that, After obtaining the optimal spectral parameters, before calculating the PWM duty cycle of each color LED, the following steps are also included: A spectral superposition equation is constructed to calculate the proportioning coefficients of each color LED; the constructed spectral superposition equation includes: Where X, Y, and Z represent the CIE 1931 standard excitation values; , , Color matching functions for CIE standard observers; , , Let be the spectral power distribution function of each LED; , , These are the proportions of red LEDs, blue LEDs, and green LEDs, respectively. λ is the wavelength of light.

3. The method according to claim 1, characterized in that, The real-time growth status images are preprocessed and feature extracted to determine the variety and current growth stage of the mushrooms, including: Adaptive histogram equalization is used to enhance the contrast of the real-time growth state image; edge noise of the real-time growth state image is removed to obtain a smooth grayscale image. The grayscale image was converted into a binary image containing mycelium and background using the Otsu dynamic threshold segmentation algorithm. Feature extraction is performed on the binary image to obtain the variety and current growth stage of the mushroom.

4. The method according to claim 3, characterized in that, Feature extraction of the binary image includes: Feature extraction can be performed on the binary image using a ResNet-18 network; The output of the ResNet-18 network is divided into three layers, which output hyphal density, primordium number, and fruiting body diameter, respectively.

5. The method according to claim 1, characterized in that, The mushroom growth model database uses structured storage. Each record in the mushroom growth model database contains a spectral stage identifier, the variety of mushroom, the growth stage of the mushroom, and the corresponding optimal spectral parameters. The spectral parameters include the red light ratio coefficient, the blue light ratio coefficient, the green light ratio coefficient, and the total target light intensity.

6. The method according to claim 1, characterized in that, Also includes: A three-layer control architecture is constructed, which includes an upper-layer main control chip, a middle-layer driver chip, and the multispectral LED light source located at the bottom layer; The upper-layer main control chip is used to determine whether the current growth stage has reached the preset conditions for triggering the stage transition. If so, it calls the mushroom growth model database, obtains the optimal spectral parameters based on the mushroom growth model database, the mushroom variety, and the current growth stage, and calculates the PWM duty cycle of each color LED based on the optimal spectral parameters. The middle layer driver chip is used to obtain the drive current signal based on the PWM duty cycle of each color LED, and to drive and adjust the multispectral LED light source based on the drive current signal.

7. The method according to claim 1, characterized in that, The multispectral LED light source also includes: Heat dissipation substrate; The multi-color LEDs are all located on the surface of the heat dissipation substrate and form a red, blue, and green three-primary-color LED array; each color LED array includes multiple driving units; each driving unit includes multiple LEDs connected in parallel; Multiple optical lenses are arranged in a one-to-one correspondence with the driving unit and are located on the light-emitting surface of each driving unit.

8. The method according to claim 6, characterized in that, The PWM duty cycle D is obtained based on the following formula: Where D is the PWM duty cycle; This refers to the ratio coefficient of the corresponding color LEDs; Total light intensity as the target; This represents the maximum luminous intensity of the corresponding color LED; This refers to the luminous efficacy attenuation coefficient of the corresponding color LED; Maximum resolution for 16-bit PWM.

9. The method according to claim 2, characterized in that, After adjusting the multispectral LED light source based on the driving current signal, the process further includes repeating the following steps several times: Real-time monitoring of the actual light signals output by each color LED; If an abnormality in the optical signal is detected, the optimal spectral parameters are corrected based on the spectral superposition equation. The PWM duty cycle is updated based on the corrected optimal spectral parameters; Update the drive current signal based on the updated PWM duty cycle; The driving adjustment of the multispectral LED light source is performed based on the updated driving current signal.

10. A mushroom growth supplemental lighting control system based on multispectral LED light source, characterized in that, The mushroom growth supplemental lighting control system based on multispectral LED light source includes: A multispectral LED light source is installed in the mushroom growth environment, and the multispectral LED light source includes multicolor LEDs; The database construction module is used to build a mushroom growth model database containing the correspondence between mushroom varieties, growth stages, and spectral parameters. The acquisition module is used to acquire real-time images of the growth status of mushrooms; The growth status recognition module is used to preprocess and extract features from the real-time growth status image to determine the mushroom variety and current growth stage. The spectral parameter calling module is used to determine whether the current growth stage has reached the preset conditions for triggering the stage transition. If so, it calls the mushroom growth model database and obtains the optimal spectral parameters based on the mushroom growth model database, the mushroom variety, and the current growth stage. A multispectral LED control module is used to calculate the PWM duty cycle of each color LED based on the optimal spectral parameters; obtain a drive current signal based on the PWM duty cycle of each color LED; and adjust the drive of the multispectral LED light source based on the drive current signal.

Images