LED plant lamp illumination system and method

Abstract

The invention discloses an LED plant lamp illumination system and method, and relates to the technical field of planting industry, and the LED plant lamp illumination system comprises a central processor which is connected with an intelligent identification module used for carrying out plant classification on data collected by a visual collection module. The shape extraction module, the color extraction module and the image preprocessing module are used in cooperation to extract effective information in image information, and then the classification identification module and the CNN deep learning network unit are used in cooperation to facilitate intelligent identification of plants. The central processing unit can intelligently control the operation of the brightness adjusting unit according to the calculation result of the intelligent identification module, and the brightness of the LED lamp can be adjusted through the power-on operation of the PWM controller and the LED controller, so that the irradiation light source of the LED lamp meets the subsequent planting and growth requirements, the function of intelligently adjusting the brightness of the LED lamp is realized, and the LED lamp is more intelligent. And the automation effect of the using device is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of planting industry, and in particular to an LED plant lamp lighting system and method. Background Art

[0002] LED plant lights can provide suitable light for plants to promote photosynthesis and thus meet the growth needs of plants. Different plants have different requirements for light during the growth process, and natural light is often difficult to fully meet their optimal growth needs. LED plant lights can provide light of specific wavelengths and intensities according to the growth characteristics of different plants, promote plant photosynthesis, and extend the lighting time. In cases where natural light is insufficient, such as in winter, rainy weather, or indoor planting environments, LED plant lights can extend the lighting time of plants, thereby increasing plant growth rate and yield.

[0003] LED plant lights are needed for some plants that require specific light cycles. Among them, the "LED lamp intelligent lighting system and method" disclosed in the publication number "CN107949117B" has solved the various drawbacks of existing LED lights, such as simple and single functions, which can no longer meet people's requirements for intelligent lighting in modern times.

[0004] However, there are still many defects in the actual use of LED plant lights with similar structures. For example, the existing technology cannot automatically adjust the brightness of the LED lamp according to different types of plants, which makes the device inconvenient for use in unattended environments. The operator is required to have rich planting experience to adjust the brightness of the LDE lamp, which increases the operator's workload and also increases the experience requirements for the operator of the plant. Therefore, it is necessary to design an LED plant lamp lighting system and method. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides an LED plant lamp lighting system and method to solve the above problems.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: An LED plant lamp lighting system includes:

[0007] A central processing unit for computing and processing data generated by the operation of the system;

[0008] Visual acquisition module for collecting data on flowers, plants and crops;

[0009] Auxiliary planting units used to intervene in the growth of flowers, plants and crops;

[0010] The central processor is connected to an intelligent recognition module for classifying plants based on data collected by the visual collection module;

[0011] The intelligent identification module is connected to a brightness adjustment unit for adjusting the brightness and wavelength of the system operation;

[0012] The auxiliary planting unit is connected to an LED lamp, and the LED lamp is connected to a brightness adjustment unit. The brightness adjustment unit includes a PWM controller and an LED driver for adjusting the brightness of the LED lamp.

[0013] Preferably, the intelligent recognition module includes a shape extraction module for performing shape extraction on information collected from flowers, plants and crops; the shape extraction module is connected to a color extraction module for extracting the color of information collected by the visual collection module; the color extraction module is connected to an image preprocessing module for preprocessing the collected image information; the image preprocessing module is connected to a classification recognition module for identifying the image information; and the classification recognition module is connected to a CNN deep learning network unit for performing deep learning on the image information.

[0014] Preferably, the visual acquisition module includes a high-resolution camera for acquiring flowers, plants and crops, the high-resolution camera is connected to an industrial camera for taking timed high-frequency photos of flowers, plants and crops, and the industrial camera is connected to a noise reduction module for performing noise reduction processing on the acquired image information.

[0015] Preferably, the auxiliary planting unit includes a macro camera for collecting data on plant texture, surface and growth trend; the macro camera is connected to an online information collection module for collecting data information on plants with different growth trends on the Internet cloud; the online information collection module is connected to a feature extraction unit for preprocessing data collected by the macro camera; the feature extraction unit is connected to a comparison module for comparing information collected by the online information collection module with information collected by the macro camera.

[0016] Preferably, the central processing unit is connected to a temperature sensor for real-time monitoring of the temperature of the device, and the temperature sensor is connected to a micro fan for dissipating heat and cooling the device.

[0017] Preferably, the central processing unit is connected to a networking unit for communicating with the Internet, the networking unit is connected to a server for storing data and logs generated by system operation, and the server is connected to a mobile terminal for network communication.

[0018] Preferably, the classification and recognition module adopts color feature extraction, and uses a color histogram to represent the color features of an image. For a color image V, its color histogram H is calculated by the following formula:

[0019]

[0020] Where v represents the color value, x and y represent the pixel coordinates in the image. By comparing the color histograms of different images, the color similarity between them can be measured;

[0021] Through shape feature extraction, the Canny edge detection is used to extract the edge of the image, and the contour of the object is obtained through the contour extraction algorithm. For a contour C, its shape features such as perimeter P and area A can be calculated. Through Roundness = 4πA / P 2 A calculation used to measure how close an object's shape is to being round.

[0022] Preferably, the CNN deep learning network unit adopts a convolutional neural network, which is composed of multiple convolutional layers, pooling layers and fully connected layers. For an input image I, after a series of convolution and pooling operations, a feature map F is obtained, and the feature map is mapped to the output category space through the fully connected layer to obtain the classification result. In the recognition process, a loss function is used to measure the difference between the prediction result of the model and the true label. Using the loss function, for a training set containing N samples, the cross entropy loss function can be expressed as: Where C is the number of categories, yij is the true label of the ith sample (if it belongs to the xth category, then yij = 1, otherwise 0), and Pij is the probability that the model predicts that the ith sample belongs to the jth category.

[0023] The present invention also discloses a method for using the LED plant lamp lighting system, the steps of which are as follows:

[0024] S1. After deploying the system to the designated location, the plant data is collected through the visual collection module;

[0025] S2. Identify the plant species through the intelligent recognition module, and adjust the light generated by the system operation through the brightness adjustment unit;

[0026] S3. The auxiliary planting unit can identify the growth status of plants at different growth stages and adjust the system illumination brightness according to the plant growth trend;

[0027] S4. By adjusting the brightness of LED lamps, light is provided for plants at different growth stages.

[0028] Beneficial Effects

[0029] The present invention provides an LED plant lamp lighting system and method. Compared with the prior art, it has the following beneficial effects:

[0030] 1. The LED plant lamp lighting system provided by the present invention extracts the plant collection data through the shape extraction module, thereby extracting the plant shape in the plant collection data, and then extracts the effective information in the image information through the color extraction module and the image preprocessing module. Then, the classification and recognition module and the CNN deep learning network unit are used together to facilitate intelligent recognition of the plants. The central processing unit can then intelligently control the operation of the brightness adjustment unit according to the calculation result of the intelligent recognition module. The brightness of the LED lamp can be adjusted by powering on the PWM controller and the LDE controller, so that the LED lamp irradiation light source meets the subsequent planting and growth requirements, and realizes the function of intelligently adjusting the brightness of the LED lamp, avoiding the need for operators to frequently adjust the brightness of the LED lamp, making the device suitable for unattended environments, increasing the automation effect of using the device, and facilitating the promotion and use of the device.

[0031] 2. The LED plant lamp lighting system provided by the present invention monitors the growth status of plants in real time through a macro camera, and collects images of plants at the same growth stage on the Internet through an online information acquisition module, and transmits them to a feature extraction unit. The feature extraction unit can extract key information from the images of plants at the same growth stage on the Internet, and transmits it to a comparison module for calculation. The comparison module can compare the images of plants with the images of plants at the same growth stage on the Internet, so as to judge the growth status of the plants. When it is judged that the growth status of the plants is not good, the brightness of the LED lamp can be adjusted by calling the brightness adjustment unit through the central processing unit, so as to facilitate additional light compensation for underdeveloped plants, realize the function of intelligent assisted planting, avoid the problem of underdeveloped plants due to insufficient planting experience of operators, and facilitate the promotion and use of the device.

[0032] 3. The LED plant lamp lighting system provided by the present invention monitors the real-time temperature of the device by powering on the temperature sensor, and transmits it to the central processing unit for calculation. When the central processing unit calculates that the temperature of the device exceeds the preset value, it controls the micro fan to power on through the wire to perform air cooling and heat dissipation on the device, thereby ensuring the long-term stable operation of the device, avoiding the problem of short circuit of the device caused by heat accumulation inside the device, and improving the reliability of the device. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the system principle of the present invention;

[0034] Figure 2 It is a schematic diagram of the visual acquisition module of the present invention;

[0035] Figure 3 It is a schematic diagram of the intelligent recognition module of the present invention;

[0036] Figure 4 It is a schematic diagram of the auxiliary planting unit of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] See also Figure 1-Figure 4 , two technical solutions are provided: The first implementation method:

[0039] An LED plant lamp lighting system comprises:

[0040] A central processing unit for computing and processing data generated by the operation of the system;

[0041] Visual acquisition module for collecting data on flowers, plants and crops;

[0042] Auxiliary planting units used to intervene in the growth of flowers, plants and crops;

[0043] The central processing unit is connected to an intelligent recognition module for classifying plants based on data collected by the visual collection module;

[0044] The intelligent identification module is connected to a brightness adjustment unit for adjusting the brightness and wavelength of the system operation;

[0045] The auxiliary planting unit is connected to an LED lamp, and the LED lamp is connected to a brightness adjustment unit, and the brightness adjustment unit includes a PWM controller and an LED driver for adjusting the brightness of the LED lamp.

[0046] The intelligent recognition module includes a shape extraction module for extracting shapes of collected information of flowers, plants and crops, the shape extraction module is connected to a color extraction module for extracting colors of information collected by the visual collection module, the color extraction module is connected to an image preprocessing module for preprocessing collected image information, the image preprocessing module is connected to a classification recognition module for recognizing image information, and the classification recognition module is connected to a CNN deep learning network unit for deep learning of image information;

[0047] The plant collection data is extracted through the shape extraction module, so as to extract the plant shape in the plant collection data. Then, the color extraction module is used in conjunction with the image preprocessing module to extract effective information from the image information. Then, the classification recognition module is used in conjunction with the CNN deep learning network unit to facilitate intelligent recognition of the plants. The central processing unit can intelligently control the operation of the brightness adjustment unit according to the calculation results of the intelligent recognition module. The brightness of the LED lamp can be adjusted by powering on the PWM controller and the LDE controller, so that the light source of the LED lamp meets the subsequent planting and growth requirements, realizing the function of intelligently adjusting the brightness of the LED lamp.

[0048] The visual acquisition module includes a high-resolution camera for collecting flowers, plants and crops, the high-resolution camera is connected to an industrial camera for taking timed high-frequency photos of flowers, plants and crops, and the industrial camera is connected to a noise reduction module for performing noise reduction processing on the collected image information;

[0049] High-resolution cameras can be used to collect plant information, including but not limited to plant texture, plant extension, plant surface spots, plant growth height, etc. Industrial cameras have the ability to capture information at different growth stages, making it easy to capture and record plant information. The noise reduction module can effectively eliminate noise in the image.

[0050] The auxiliary planting unit includes a macro camera for collecting data on plant texture, surface, and growth trend. The macro camera is connected to an online information collection module for collecting data information on plants with different growth trends on the Internet cloud. The online information collection module is connected to a feature extraction unit for preprocessing data collected by the macro camera. The feature extraction unit is connected to a comparison module for comparing information collected by the online information collection module with information collected by the macro camera.

[0051] The growth status of the plants is monitored in real time through a macro camera, and the images of the plants at the same growth stage on the Internet are collected through an online information collection module and transmitted to the feature extraction unit. The feature extraction unit can extract key information from the images of the plants at the same growth stage on the Internet and transmit it to the comparison module for calculation. The comparison module can compare the images of the plants with the images of the plants at the same growth stage on the Internet, so as to judge the growth status of the plants. When it is judged that the growth status of the plants is not good, the brightness of the LED lamps can be adjusted by calling the brightness adjustment unit through the central processor, so as to facilitate additional light compensation for underdeveloped plants, thereby realizing the function of intelligent auxiliary planting;

[0052] The central processing unit is connected to a networking unit for communicating with the Internet, the networking unit is connected to a server for storing data and logs generated by system operation, and the server is connected to a mobile terminal for network communication;

[0053] The mobile terminal can be used to remotely understand the operating status of the device, and the networking unit can facilitate the communication between the device and the Internet;

[0054] The classification and recognition module uses color feature extraction and uses color histogram to represent the color features of the image. For a color image V, its color histogram H is calculated by the following formula:

[0055]

[0056] Where v represents the color value, x and y represent the pixel coordinates in the image. By comparing the color histograms of different images, the color similarity between them can be measured;

[0057] Through shape feature extraction, the Canny edge detection is used to extract the edge of the image, and the contour of the object is obtained through the contour extraction algorithm. For a contour C, its shape features such as perimeter P and area A can be calculated. Through Roundness = 4πA / P 2 Calculation, used to measure the degree to which the shape of the object is close to a circle, compares the extracted features with the feature library of known plant species deployed inside the classification and recognition module, and uses distance metrics, Euclidean distance, and cosine similarity to determine the most similar plant species;

[0058] The CNN deep learning network unit uses a convolutional neural network, which consists of multiple convolutional layers, pooling layers, and fully connected layers. For an input image I, after a series of convolution and pooling operations, a feature map F is obtained. The feature map is mapped to the output category space through the fully connected layer to obtain the classification result. In the recognition process, the loss function is used to measure the difference between the model's prediction result and the true label. Using the loss function, for a training set containing N samples, the cross entropy loss function can be expressed as: Where C is the number of categories, yij is the true label of the ith sample (if it belongs to the xth category, yij=1, otherwise 0), Pij is the probability that the model predicts that the ith sample belongs to the jth category. The parameters of the model are continuously adjusted through the back propagation algorithm and optimization algorithm (such as stochastic gradient descent) to minimize the loss function, thereby improving the classification accuracy of the recognition model.

[0059] The second implementation mode is mainly different from the first implementation mode in that: the central processing unit is connected to a temperature sensor for real-time monitoring of the temperature of the device, and the temperature sensor is connected to a micro fan for dissipating heat and cooling the device.

[0060] The temperature sensor is powered on to monitor the real-time temperature of the device and transmits it to the central processing unit for calculation. When the central processing unit calculates that the temperature of the device exceeds the preset value, the micro fan is controlled by the wire to power on and cool the device, ensuring the stable operation of the device for a long time.

[0061] The present invention also discloses a method for using the LED plant lamp lighting system, the steps of which are as follows:

[0062] S1. After the system is deployed at the designated location, the high-resolution camera can be used to collect plant information, including but not limited to plant texture, plant extension, plant surface spots, plant growth height, etc. The noise reduction module can effectively eliminate noise points in the image;

[0063] S2. The plant collection data is extracted through the shape extraction module, so as to extract the plant shape in the plant collection data, and then the color extraction module is used in conjunction with the image preprocessing module to extract effective information from the image information, and then the classification recognition module is used in conjunction with the CNN deep learning network unit to facilitate intelligent recognition of the plants. The central processing unit can then intelligently control the operation of the brightness adjustment unit according to the calculation results of the intelligent recognition module, so that the LED lamp irradiation light source meets the subsequent planting growth requirements;

[0064] S3. Real-time monitoring of the growth status of the plants is performed through a macro camera, and the images of the plants at the same growth stage on the Internet are collected through the networked information collection module, and transmitted to the feature extraction unit. The feature extraction unit can extract key information from the images of the plants at the same growth stage on the Internet, and transmit it to the comparison module for calculation. The comparison module can compare the images of the plants with the images of the plants at the same growth stage on the Internet, and the operation of the intelligent control brightness adjustment unit can adjust the brightness of the LED lamps, thereby realizing the function of intelligent auxiliary planting;

[0065] S4. By adjusting the brightness of LED lamps, you can provide light for plants at different growth stages. In the seedling stage, plants are usually delicate and have relatively low requirements for light intensity. Excessive light may cause damage to seedlings, such as burning leaves and inhibiting growth. As plants grow, the demand for light intensity gradually increases. During the growth period, plants need sufficient light to carry out photosynthesis, synthesize enough organic matter, and promote plant growth and development. During the flowering and fruiting periods, plants have relatively high requirements for light intensity and light time. Adequate light can promote flower bud differentiation, flower opening and fruit ripening. At this time, the brightness of LED lamps can be further improved, and the light time can be appropriately extended to provide plants with more sufficient light.

[0066] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0067] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An LED plant lamp lighting system, characterized in that: include: A central processing unit for computing and processing data generated by the operation of the system; Visual acquisition module for data collection of flowers, plants and crops; Auxiliary planting units used to intervene in the growth of flowers, plants and crops; The central processor is connected to an intelligent recognition module for classifying plants based on data collected by the visual collection module; The intelligent identification module is connected to a brightness adjustment unit for adjusting the brightness and wavelength of the system operation; The auxiliary planting unit is connected to an LED lamp, and the LED lamp is connected to a brightness adjustment unit. The brightness adjustment unit includes a PWM controller and an LED driver for adjusting the brightness of the LED lamp.

2. The LED plant lamp lighting system according to claim 1, characterized in that: The intelligent recognition module includes a shape extraction module for extracting shapes of information collected from flowers, plants and crops; the shape extraction module is connected to a color extraction module for extracting the color of information collected by the visual collection module; the color extraction module is connected to an image preprocessing module for preprocessing the collected image information; the image preprocessing module is connected to a classification recognition module for identifying image information; and the classification recognition module is connected to a CNN deep learning network unit for deep learning of image information.

3. The LED plant lamp lighting system according to claim 1, characterized in that: The visual acquisition module includes a high-resolution camera for acquiring flowers, plants and crops, the high-resolution camera is connected to an industrial camera for taking timed high-frequency photos of flowers, plants and crops, and the industrial camera is connected to a noise reduction module for performing noise reduction processing on the acquired image information.

4. The LED plant lamp lighting system according to claim 1, characterized in that: The auxiliary planting unit includes a macro camera for collecting data on plant texture, surface and growth trend; the macro camera is connected to an online information collection module for collecting data information on plants with different growth trends on the Internet cloud; the online information collection module is connected to a feature extraction unit for preprocessing data collected by the macro camera; the feature extraction unit is connected to a comparison module for comparing information collected by the online information collection module with information collected by the macro camera.

5. The LED plant lamp lighting system according to claim 1, characterized in that: The central processor is connected to a temperature sensor for real-time monitoring of the temperature of the device, and the temperature sensor is connected to a micro fan for dissipating heat and cooling the device.

6. The LED plant lamp lighting system according to claim 1, characterized in that: The central processor is connected to a networking unit for communicating with the Internet, the networking unit is connected to a server for storing data and logs generated by system operation, and the server is connected to a mobile terminal for network communication.

7. The LED plant lamp lighting system and method according to claim 2, characterized in that: The classification and recognition module adopts color feature extraction and uses color histogram to represent the color features of the image. For a color image V, its color histogram H is calculated by the following formula: Where v represents the color value, x and y represent the pixel coordinates in the image, and the color similarity between different images is measured by comparing their color histograms; Through shape feature extraction, the Canny edge detection is used to extract the edge of the image, and the contour of the object is obtained through the contour extraction algorithm. For a contour C, its perimeter P and area A shape features are calculated, and Roundness = 4πA / P 2 A calculation used to measure how close an object's shape is to being round.

8. The LED plant lamp lighting system according to claim 2, characterized in that: The CNN deep learning network unit adopts a convolutional neural network, which consists of multiple convolutional layers, pooling layers and fully connected layers. For an input image I, after a series of convolution and pooling operations, a feature map F is obtained, and the feature map is mapped to the output category space through the fully connected layer to obtain the classification result. In the recognition process, a loss function is used to measure the difference between the model's prediction result and the true label. Using the loss function, for a training set containing N samples, the cross entropy loss function is expressed as: Where C is the number of categories, yij is the true label of the i-th sample, and Pij is the probability that the model predicts that the i-th sample belongs to the j-th category.

9. A method for using an LED plant lamp lighting system, characterized in that: Using the LED plant lamp lighting system according to any one of claims 1 to 9, the steps are as follows: S1. After deploying the system to the designated location, the plant data is collected through the visual collection module; S2. Identify the plant species through the intelligent recognition module, and adjust the light generated by the system operation through the brightness adjustment unit; S3. The auxiliary planting unit can identify the growth status of plants at different growth stages and adjust the system illumination brightness according to the plant growth trend; S4. By adjusting the brightness of LED lamps, light is provided for plants at different growth stages.

Images