Lighting method and system based on spectral wavelength, electronic equipment and medium

Abstract

The invention discloses an illumination method and system based on spectral wavelength, electronic equipment and a medium, and relates to the technical field of intelligent illumination. The method comprises the following steps: acquiring a chicken flock image of a target breeding area; according to an image recognition algorithm, determining a chicken flock distribution position and ambient light brightness in the chicken flock image; inputting the chicken flock distribution position and the ambient light brightness into a preset spectral wavelength prediction model to obtain a spectral wavelength prediction result; and according to the spectral wavelength prediction result, generating a light control scheme, and according to the light control scheme, controlling the light of the breeding area for illumination. Through the scheme, the chicken flock distribution position and the ambient light brightness are determined according to the chicken flock image, adaptive spectral wavelength adjustment is performed according to the chicken flock distribution position and the ambient light brightness, and the effects of spectral wavelength adjustment and auxiliary illumination are achieved according to the chicken flock distribution position and the ambient light brightness.

Description

Technical Field

[0001] The present application relates to the field of intelligent lighting technology, and in particular to a lighting method, system, electronic device and medium based on spectral wavelength. Background Art

[0002] With the continuous development of agriculture and poultry farming, the capture of chickens has become an important requirement for improving farming efficiency and management. When the chickens are on the shelves to rest, catching them becomes more challenging. In order to solve this problem, lighting and other methods can be used to assist in finding the target chickens to be caught, so as to achieve fixed-point capture.

[0003] At present, when chicken house workers need to capture targets in the flock, they often use fixed-wavelength lighting methods such as flashlights or lighting lamps. However, in actual applications, since the chickens in the flock are in different states, the traditional fixed-wavelength lighting method will cause irritation to the chickens in the resting state, causing the chickens to become agitated, making it more difficult to capture. Summary of the invention

[0004] The present application provides a lighting method, system, electronic device and medium based on spectral wavelength, which can adjust the spectral wavelength according to the distribution position of chickens and the brightness of ambient light, thereby achieving the effect of auxiliary lighting.

[0005] In a first aspect, the present application provides a spectral wavelength-based lighting method, comprising: Acquire images of chicken flocks in the target breeding area; Determine the distribution position of the chickens and the ambient light brightness in the chicken image according to the image recognition algorithm; Input the chicken distribution position and the ambient light brightness into a preset spectral wavelength prediction model to obtain a spectral wavelength prediction result; A lighting control scheme is generated according to the spectral wavelength prediction result, and the lighting in the breeding area is controlled according to the lighting control scheme for illumination.

[0006] By adopting the above technical solution, the image is processed, the distribution position of the chickens in the target area where the chickens are located is determined from the edge pixel coordinates in the position feature, and the ambient light brightness of the current scene is evaluated from the pixel grayscale value in the illumination feature. This analysis based on objective image data avoids the shortcomings of manual subjective judgment and obtains information more accurately and reliably. After the extracted chicken distribution position and ambient light brightness information are input into the preset spectral wavelength prediction model, it can intelligently predict which spectral wavelength combination and intensity should be used for lighting in this scene. The spectral wavelength prediction model is obtained by training with a machine learning algorithm, and can output a spectral wavelength adjustment scheme according to different chicken states and environmental conditions. According to the spectral wavelength prediction results, a corresponding lighting control scheme is generated to control the lighting in the breeding area for lighting. By adjusting the spectral combination and intensity, a suitable lighting environment can be created to reduce the stimulation to the chickens, avoid the harassment that may be caused by traditional fixed wavelength lighting, and is conducive to capturing targets while reducing the impact on other chickens, and providing auxiliary lighting for staff.

[0007] Optionally, an initial image of the chicken flock in the target breeding area is obtained by an infrared camera; temperature information of the target breeding area is obtained by a temperature sensor according to a preset acquisition frequency; and the initial image of the chicken flock and the temperature information are converted into a corresponding image of the chicken flock in the target breeding area according to a preset coding format.

[0008] By adopting the above technical solution, the initial image of the chicken flock is obtained according to the preset acquisition frequency, and the temperature data of the environment in which the chicken flock is located is collected regularly. The initial image and the temperature data are fused according to the preset coding format, such as writing the corresponding temperature information at different positions of the image. The image after coding is the image of the chicken flock in the target breeding area, which contains both the visual information and temperature data of the chicken flock. This technical solution combines visual images with temperature parameters. Temperature information can help to more accurately judge the active state of the chicken flock. The coding format ensures the integration of information and avoids the complexity of subsequent algorithm processing. The above derivation realizes the simultaneous acquisition of temperature information when acquiring the image data of the chicken flock, as well as the integrated coding of multi-source data, and obtains richer image data, which provides assistance for subsequent state analysis and judgment, and improves the accuracy of judgment.

[0009] Optionally, the image recognition algorithm is a convolutional neural network model, which is obtained by cyclic training of images of chicken flocks in historical breeding areas, image features of chicken flock positions, and ambient light image features.

[0010] By adopting the above technical solutions, the convolutional neural network can effectively extract key feature information such as the position of the chicken flock and the ambient light from the complex chicken flock image scene. Compared with the traditional manual feature extraction method, the convolutional neural network can automatically learn the hierarchical pattern of the image and gradually extract the feature representation from low-level to high-level, avoiding the difficulties and limitations of manually designed features. The practice of using a large amount of historical data to perform cyclic training on the convolutional neural network model enables the model to continuously learn and optimize the ability of feature extraction during the training process. By using the chicken flock images, position features and lighting feature data of various typical scenes in the historical breeding area, the model can continuously summarize and refine the laws and experience of the chicken flock position distribution and the environmental brightness assessment, thereby showing a strong generalization ability in practical applications. The trained convolutional neural network model can efficiently and accurately extract the position features of the chicken flock distribution and the features of the ambient light from the input image, providing an accurate decision-making basis for the subsequent generation of lighting control solutions. This method based on image data analysis can avoid judgment errors and improve the accuracy of position determination and brightness assessment.

[0011] Optionally, the convolutional neural network model is used to extract position features and illumination features in the image of the chicken flock in the target breeding area; the distribution position of the chicken flock is determined based on the coordinates of the edge pixel points in the position features; and the ambient light brightness is determined based on the pixel grayscale value in the illumination features.

[0012] By adopting the above technical solution, the convolutional neural network is used to automatically extract position features and lighting features from complex images, which is more efficient and accurate than manual feature extraction. The edge pixel coordinates in the position feature can accurately reflect the distribution position of the chickens in the breeding area, and the pixel grayscale value in the lighting feature can objectively evaluate the brightness of the current scene. These key information will provide a decision-making basis for the subsequent generation of targeted lighting control solutions. By determining the position of the chickens through the edge pixel coordinates, it is possible to avoid lighting the entire breeding area, thereby saving energy. At the same time, local lighting can be implemented in the target area where the chickens are located to reduce the impact and disturbance on chickens in other areas. If a specific chicken needs to be captured, its position can be used as the lighting center for fixed-point lighting. The pixel grayscale value is used to evaluate the ambient brightness, which can effectively avoid the defects of manual subjective judgment and obtain the real-time lighting conditions on the scene. Adjusting the light brightness and spectral intensity according to different brightness scenes can create a suitable lighting environment and avoid excessive or missing lighting.

[0013] Optionally, historical chicken distribution positions and historical ambient light brightness are obtained, and the historical chicken distribution positions and the historical ambient light brightness are used as historical evaluation indicators; historical climate information corresponding to the historical evaluation indicators is determined, and the historical climate information includes cloudy days, sunny days and thunderstorm days; the actual spectral wavelength is determined according to the historical evaluation indicators and the historical climate information corresponding to the historical evaluation indicators; the historical evaluation indicators, the historical climate information corresponding to the historical evaluation indicators, and the actual spectral wavelength are input into the initial neural network model to obtain the target spectral wavelength prediction result; the initial neural network model is iteratively trained according to the target spectral wavelength prediction result to obtain the preset spectral wavelength prediction model.

[0014] By adopting the above technical solution, historical data including different chicken distribution positions, ambient light brightness and corresponding climatic conditions are collected, and the illumination spectrum parameters used at that time are determined as historical spectral data based on these data. The initial spectral prediction model is established by using a neural network algorithm. The collected historical data including status and climate are input to obtain the spectral prediction results, and the difference between the prediction results and the actual spectral data is calculated. If the difference exceeds the preset threshold, the network model is continuously trained for iterative optimization until the network prediction spectrum reaches a satisfactory accuracy and a trained spectral prediction model is obtained. When the new chicken distribution position, ambient light brightness and climate data are input, the corresponding suitable illumination spectrum results can be predicted. This technical solution uses big data to train a spectral prediction model, which can automatically learn the mapping relationship between status, climate and illumination parameters. Iterative training makes the model more accurate, which provides a guarantee for the subsequent intelligent adjustment of the spectral wavelength according to the real-time position of the chickens and the ambient light brightness.

[0015] Optionally, a weighted sum is performed on the historical chicken distribution positions and historical ambient light brightness in the historical evaluation indicators to obtain evaluation values ​​of the chicken distribution positions and ambient light brightness; if the chicken distribution positions and ambient light brightness evaluation values ​​are greater than or equal to a state evaluation value threshold, the proportion of red light and orange light in the spectral wavelength is increased; if the chicken distribution positions and ambient light brightness evaluation values ​​are less than a state evaluation value threshold, the proportion of blue light and white light in the spectral wavelength is increased; according to a historical climate information mapping table, the proportion values ​​of the blue light, the white light, the red light and the orange light in the spectral wavelength corresponding to cloudy days, sunny days and thunderstorm days are determined; the actual spectral wavelength is obtained according to the corresponding proportion values ​​of the blue light, white light, red light and orange light in the spectral wavelength.

[0016] By adopting the above technical solution, according to the collected historical data, the spectrum usage rules under different chicken distribution positions and ambient light brightness are analyzed, and the conclusion is drawn that the proportion of red light and orange light is enhanced when the chickens are active, and the proportion of blue light and white light is enhanced when the chickens are inactive. A table of climate conditions and spectrum mapping relationships is established to determine the proportion values ​​of various spectra under different chicken distribution positions and ambient light brightness under different cloudy, sunny and thunderstorm climates. Input the current chicken distribution position and ambient light brightness judgment results and actual climate conditions, refer to the above-mentioned state spectrum proportion rules and climate mapping relationship, and calculate the current proportion values ​​of blue light, white light, red light, and orange light as the actual spectrum wavelength results. This technical solution can determine the appropriate spectrum wavelength proportion according to the actual chicken distribution position, ambient light brightness and climate change conditions to avoid irritation to the chickens. Determine the spectrum proportion according to the state in different climates, realize accurate lighting parameter control, comprehensively consider state adjustment and environmental influences, and provide data support for subsequent lighting control.

[0017] Optionally, image information of target chickens to be captured is obtained; based on the image information of the target chickens to be captured, location information of the target chickens to be captured in the target breeding area is determined; based on the location information, a lighting range for the target captured chickens is determined; based on the lighting range, a lighting angle adjustment instruction is generated and sent to a lighting control terminal in the target breeding area to adjust the lighting angle to illuminate the target captured chickens.

[0018] By adopting the above technical solution, the image information of the target chicken to be caught is obtained, and the image analysis technology is used to determine the precise position information of the target chicken to be caught in the breeding area, which reflects the movement trajectory and direction of the target chicken to be caught. Based on the position information of the target chicken to be caught, its corresponding illumination range, that is, the area that needs to be illuminated, can be calculated. Then, by generating a light angle adjustment instruction, the lighting direction in the breeding area is controlled in real time, so that the lighting range can be adjusted according to the movement of the target chicken to be caught, ensuring that the light is always focused on the target chicken to be caught, and accurate tracking lighting is performed.

[0019] Optionally, based on the target arrested chicken image information, the posture characteristics and behavioral characteristics of the target arrested chicken are extracted; according to the posture characteristics and the behavioral characteristics, the current resting state of the target arrested chicken is determined in a preset state feature mapping table; combined with the current resting state and position information of the target arrested chicken, prompt information is generated, and the prompt information is sent to the chicken catching personnel terminal.

[0020] By adopting the above technical solution, by analyzing and extracting the posture characteristics and behavioral characteristics of the target chicken to be caught, such as the head rotation frequency, gait parameters, etc., and searching for matches in the preset state feature mapping table, the current resting state of the target chicken to be caught can be intelligently determined, such as alert, walking, eating, etc. In combination with the resting state of the target chicken to be caught and the determined precise location information, highly targeted capture prompt information can be generated and sent to the chicken catcher. Intelligent analysis and judgment of the chicken state is realized, avoiding the judgment errors caused by relying solely on manual experience. The location information combined with the specific resting state of the target chicken to be caught enables the chicken catcher to grasp the optimal capture time and improve the capture efficiency.

[0021] In a second aspect of the present application, a system for a spectral wavelength-based lighting method is provided, the system comprising: An information acquisition module is used to obtain images of chicken flocks in a target breeding area; An image recognition module, used to determine the distribution position of the chickens and the ambient light brightness in the chicken image according to an image recognition algorithm; A model prediction module, used to input the chicken distribution position and the ambient light brightness into a preset spectral wavelength prediction model to obtain a spectral wavelength prediction result; The lighting control module is used to generate a lighting control plan according to the spectral wavelength prediction result, and control the lighting in the breeding area for illumination according to the lighting control plan.

[0022] In a third aspect of the present application, an electronic device is provided, comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein the program can implement a lighting method based on spectral wavelength when loaded and executed by the processor.

[0023] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: By adopting the above technical solution, the image is processed, the distribution position of the chickens in the target area where the chickens are located is determined from the edge pixel coordinates in the position feature, and the ambient light brightness of the current scene is evaluated from the pixel grayscale value in the illumination feature. This analysis based on objective image data avoids the shortcomings of manual subjective judgment and obtains information more accurately and reliably. After the extracted chicken distribution position and ambient light brightness information are input into the preset spectral wavelength prediction model, it can intelligently predict which spectral wavelength combination and intensity should be used for lighting in this scene. The spectral wavelength prediction model is obtained by training with a machine learning algorithm, and can output a spectral wavelength adjustment scheme according to different chicken states and environmental conditions. According to the spectral wavelength prediction results, a corresponding lighting control scheme is generated to control the lighting in the breeding area for lighting. By adjusting the spectral combination and intensity, a suitable lighting environment can be created to reduce the stimulation to the chickens, avoid the harassment that may be caused by traditional fixed wavelength lighting, and is conducive to capturing targets while reducing the impact on other chickens, and providing auxiliary lighting for staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flow chart of a lighting method based on spectral wavelength provided in an embodiment of the present application.

[0025] Figure 2 It is a schematic diagram of chicken distribution position and ambient light brightness identification based on a spectral wavelength lighting method provided in an embodiment of the present application.

[0026] Figure 3 It is a structural schematic diagram of a spectral wavelength-based lighting system provided in an embodiment of the present application.

[0027] Figure 4 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application.

[0028] Explanation of the reference numerals: 301, information acquisition module; 302, image recognition module; 303, model prediction module; 304, lighting control module; 400, electronic device; 401, processor; 402, memory; 403, user interface; 404, network interface; 405, communication bus. DETAILED DESCRIPTION

[0029] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0030] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0031] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0032] In order to facilitate understanding of the method and system provided by the embodiments of the present application, before introducing the embodiments of the present application, the background of the embodiments of the present application is first introduced.

[0033] At present, when chicken house workers need to capture targets in the flock, they often use fixed bright lights such as flashlights or lighting to assist in lighting. However, in actual applications, the traditional fixed bright lighting method will cause irritation to the chickens in a resting state, causing the chickens to become agitated, making it more difficult to capture.

[0034] The embodiment of the present application discloses a lighting method based on spectral wavelength, which obtains images of chickens in the target breeding area through infrared camera equipment, and identifies and processes the images of chickens through image recognition algorithm, i.e., convolutional neural network model, determines the distribution position of chickens and the brightness of ambient light, inputs the distribution position of chickens and the brightness of ambient light into a preset spectral wavelength prediction model, obtains the spectral wavelength prediction result, and generates a lighting control scheme corresponding to the spectral wavelength prediction result to control the light for adaptive lighting. It is mainly used to solve the problem that the traditional fixed bright light lighting method is used for lighting. Since the chickens in the chicken flock are in different states, it will cause irritation to the chickens in the resting state, causing a commotion among the chickens and increasing the difficulty of catching them.

[0035] After the above background content is introduced, those skilled in the art can understand the problems existing in the prior art. The technical solutions in the embodiments of the present application will be described in detail below in conjunction with the drawings in the embodiments of the present application. The described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0036] Reference Figure 1 , a lighting method based on spectral wavelength, the method includes S10 to S40, specifically including the following steps: S10: Acquire images of chicken flocks in the target breeding area.

[0037] The target breeding area is the designated breeding area where the chicken catching operation is planned, for example, the designated breeding area is one of a small chicken house, a medium chicken house or a large chicken house. The target breeding area is dynamically changed and is determined each time according to the actual chicken catching needs.

[0038] Specifically, in order to monitor and judge the status and behavior of the chickens in real time, it is necessary to obtain the image information of the chickens in the target breeding area. An infrared camera installed on the top of the chicken house is used to continuously shoot and obtain the initial image of the chickens at a bird's-eye view, because the infrared camera can work in a low-light environment at night and will not interfere with the chickens. At the same time, a temperature sensor configured in the chicken house is used to detect the ambient temperature of the breeding area in real time at a certain frequency (such as once every 10 seconds). Then, the initial image of the chickens obtained by the infrared camera and the temperature information collected by the temperature sensor are packaged into target chicken image data according to a preset encoding format, such as XML or JSON. The target chicken image contains both dynamic image information of the chickens and ambient temperature information, which can more comprehensively reflect the current activity status of the chickens and the surrounding environment. This provides data support for the subsequent use of image recognition models to judge the specific status of the chickens, and to determine the lighting parameters according to the distribution location of the chickens, the ambient light brightness and the ambient temperature, so as to achieve on-demand adjustment of lighting and assist in the capture of specific chickens at night.

[0039] S20: Determine the distribution position of the chickens and the ambient light brightness in the chicken image according to the image recognition algorithm.

[0040] Among them, the image recognition algorithm is an algorithm model used to analyze the image of the chicken flock and determine the distribution position of the chicken flock and the brightness of the ambient light. The image recognition algorithm adopted by this technical solution is a convolutional neural network model. The convolutional neural network is obtained by collecting a large number of chicken flock images in historical breeding areas for training. These historical chicken flock images contain both the image features of the marked chicken flock distribution and the image features reflecting the ambient light conditions. These labeled chicken flock distribution positions and ambient light brightness sample images are repeatedly input into the convolutional neural network for training, so that the network gradually learns to distinguish and extract the characteristic information of the chicken flock images under different position distributions and ambient light brightness. After continuous cyclic training, the convolutional neural network model can accurately identify the chicken flock images and determine their chicken flock distribution positions and ambient light brightness. The convolutional neural network model obtained by such training is the image recognition algorithm adopted by this technical solution, which is used to analyze the chicken flock images in the target breeding area and determine the chicken flock distribution positions and ambient light brightness.

[0041] The distribution location of the chicken flock refers to the specific location distribution of the chicken flock in the breeding area. In the embodiments of the present application, it can be understood that the location features extracted from the chicken flock image using an image recognition algorithm (such as a convolutional neural network model), such as the edge contour coordinates of the chicken flock, are used to determine the distribution area of ​​the chicken flock in the breeding area. The chicken flock distribution location information is used to determine the target area that needs to be illuminated.

[0042] Ambient light brightness refers to the current light brightness level in the breeding area. In the embodiments of the present application, it can be understood that the lighting features extracted from the chicken image using the image recognition algorithm, such as the pixel grayscale value of the image, are used to evaluate the brightness of the current environment. The ambient light brightness information will be input into the spectral wavelength prediction model together with the chicken distribution position to determine the appropriate lighting spectrum combination.

[0043] Specifically, an initial image of a flock of chickens in the target breeding area is obtained by means of infrared cameras or the like. The initial image and the temperature information obtained by the temperature sensor are then converted into a corresponding image of a flock of chickens in the target breeding area according to a preset coding format. Next, the convolutional neural network model in the image recognition algorithm is used to process and extract features from the flock of chickens image. The convolutional neural network model is trained using images of flocks of chickens in historical breeding areas, features of flocks of chickens position images, and features of ambient light images. During the feature extraction process, the convolutional neural network model extracts position features and illumination features from the flock of chickens image. According to the edge pixel coordinates in the position features, the current distribution position of the flock of chickens in the breeding area can be determined; according to the pixel grayscale value in the illumination features, the brightness level of the current environment can be evaluated. The specific steps include S21 to S23: S21: Extract the location features and lighting features from the chicken images in the target breeding area through a convolutional neural network model.

[0044] Among them, the position feature refers to the feature in the image that reflects the position distribution information of the chicken flock. In the embodiment of the present application, it can be understood that the convolutional neural network model extracts the chicken flock outline, edge coordinates and other features from the chicken flock image to determine the specific distribution position of the chicken flock in the breeding area.

[0045] Lighting features refer to features in an image that reflect scene lighting condition information. In the embodiments of the present application, it can be understood that the convolutional neural network model extracts pixel brightness values, contrast and other features from the chicken image to evaluate the light brightness level of the current environment.

[0046] Specifically, the acquired images of chickens in the target breeding area are input into a pre-trained convolutional neural network model. The convolutional neural network model is obtained by cyclic training with a large number of images of chickens in historical breeding areas, features of chickens’ positions and features of ambient light images, and has learned how to extract features such as chickens’ positions and lighting conditions from complex images. During the extraction process, the convolutional neural network performs multi-level convolution and pooling operations on the input image, automatically learns the hierarchical pattern of the image, and gradually extracts features from low-level to high-level. Finally, the model outputs two types of feature vectors, one is the position feature vector, which contains position information such as chickens’ outlines and edges; the other is the lighting feature vector, which contains lighting condition information such as image brightness and contrast. Through this process, the specific regional location of the chickens’ distribution can be accurately extracted from the complex images of chickens, and the light brightness level of the current scene can be evaluated at the same time.

[0047] S22: Determine the distribution position of the chicken group according to the edge pixel coordinates in the position feature.

[0048] The edge pixel coordinates refer to the coordinate values ​​of the pixel points representing the target object, i.e., the boundary contour of the flock of chickens, in the image coordinate system. In the embodiment of the present application, it can be understood that the position feature vector extracted from the flock of chickens image by the convolutional neural network model contains the coordinate values ​​of a series of pixel points reflecting the edge contour of the flock of chickens.

[0049] Specifically, when the convolutional neural network model extracts features from the image of the chicken flock, it will output a position feature vector containing the position information of the chicken flock's outline, edge, etc. In this position feature vector, the coordinate values ​​of the edge pixel points can reflect the distribution area of ​​the chicken flock on the image plane. The system will process these edge pixel coordinates and map them to the actual three-dimensional breeding area coordinate system to obtain the coordinate range of one or more rectangular areas, that is, determine the distribution position of the chicken flock in the breeding area.

[0050] S23: Determine the ambient light brightness according to the pixel grayscale value in the illumination feature.

[0051] Among them, the pixel gray value refers to the brightness value of each pixel in the image, which is usually expressed as an unsigned integer and ranges from 0 (completely black) to 255 (completely white). In the embodiment of the present application, it can be understood that the illumination feature vector extracted from the chicken image by the convolutional neural network model contains the gray value corresponding to each pixel of the image.

[0052] Specifically, when the convolutional neural network model extracts features from the chicken image, it outputs a lighting feature vector containing lighting information such as image brightness and contrast. Among them, the pixel grayscale value reflects the brightness of each pixel in the image, which can more intuitively reflect the lighting level of the entire image scene. The system will perform statistics and analysis on the pixel grayscale values ​​in the lighting feature vector, such as calculating the average grayscale value, grayscale value distribution histogram, etc. By comparing with the pre-set brightness threshold or standard grayscale value, it can be judged whether the current environment is bright, dim or normal lighting state.

[0053] S30: Input the chicken distribution position and the ambient light brightness into a preset spectral wavelength prediction model to obtain a spectral wavelength prediction result.

[0054] Among them, the preset spectral wavelength prediction model refers to a pre-established model for predicting the illumination spectrum parameters corresponding to different chicken distribution positions and ambient light brightness, and the preset spectral wavelength prediction model is obtained by training a neural network model.

[0055] Specifically, a spectral wavelength prediction model based on a neural network is constructed and trained using a large amount of historical data to learn the lighting patterns under different chicken distribution positions and ambient light brightness. The input variables are the chicken distribution position and ambient light brightness, and the output result is the spectral wavelength prediction result. After determining the current chicken distribution position and ambient light brightness, the distribution position and ambient light brightness are input into the pre-trained spectral prediction model. The spectral prediction model outputs a predicted spectral wavelength result based on the input chicken distribution position and ambient light brightness, combined with the learned state and light mapping relationship.

[0056] On the basis of the above embodiment, there is also a training step of a preset spectral wavelength prediction model, and the specific process includes: continuously collecting the chicken distribution position and ambient light brightness and the corresponding actual illumination wavelength parameters in different periods and different climatic conditions in multiple chicken farms as historical data. Climate conditions include cloudy days, sunny days and thunderstorms. An initial spectral wavelength prediction model is built using intelligent algorithms such as neural networks. The model inputs the historical chicken distribution position, ambient light brightness and climate data, and outputs the predicted spectral wavelength results. The collected historical state data, i.e., the historical chicken distribution position and the corresponding historical ambient light brightness, climate data and the corresponding actual spectral data are input into the initial model for training, so that it gradually learns the mapping relationship between the two. In order to evaluate the model effect, the spectral results predicted by the trained model are compared with the actual spectral data, and the difference between the two is calculated. If the difference is greater than the preset threshold, the model is continued to be trained for iterative optimization. After continuous training, when the difference between the model prediction result and the actual spectrum is less than the threshold, it means that the training effect is satisfactory. The model finally obtained is the required spectral wavelength prediction model. The model obtained through iterative training can achieve high-precision spectral wavelength prediction based on the distribution of chickens, ambient light brightness and climatic conditions, thereby guiding the control of lighting equipment and improving chicken catching efficiency.

[0057] In an optional embodiment of the present application, the process of obtaining the actual spectrum wavelength includes: according to the collected historical data, analyzing the spectrum usage rules under different chicken distribution positions and ambient light brightness, and obtaining the conclusion that when the chickens are in an active state, that is, when the chicken state evaluation value is greater than or equal to the state evaluation value threshold, the proportion of red light and orange light is increased, and when the chickens are in an inactive state, that is, when the chicken state evaluation value is less than the state evaluation value threshold, the proportion of blue light and white light is increased. A mapping relationship table between climate conditions and spectrum proportions is formulated, and the proportion values ​​of various spectra in active and inactive states are determined under different climates such as cloudy days, sunny days, and thunderstorms. The current chicken distribution position, ambient light brightness, and climate conditions are input, and the proportion values ​​of blue light, white light, red light, and orange light in the spectrum wavelength in the current scene are calculated according to the state proportion law and the climate mapping relationship table. In this way, the lighting parameters can be determined according to the actual situation, so that the chickens can maintain a suitable active state, and the influence of different climates can be considered to achieve precise control of the light wavelength, effectively improving the efficiency of catching chickens.

[0058] In another optional embodiment of the present application, the process of analyzing the spectrum usage rules under different chicken distribution positions and ambient light brightness specifically includes: collecting a large amount of historical spectrum usage data under different chicken distribution positions and ambient light brightness. The data is sorted and sorted according to the active and inactive states of the chickens. Statistical analysis is performed on the spectral data under each state, and the proportion of different spectral components such as red light and blue light in the total spectrum is calculated. Visual analysis is performed on the distribution of the proportion of each spectral component under different states. Data analysis methods such as regression analysis are used to evaluate the correlation between each spectral component and the chicken distribution position and ambient light brightness. Based on the analysis results of the spectral data sets of each state, it is concluded that the enhancement of red light and orange light can boost vitality when the chickens are active, and blue light and white light can help sleep when they are inactive. Verification is carried out in multiple farms to ensure that the spectrum usage rules are universal. Determine the rules for increasing the proportion of red light, orange light or blue light and white light under different chicken distribution positions and ambient light brightness, and provide a basis for the subsequent determination of the proportion of spectral wavelengths.

[0059] S40: Generate a lighting control plan based on the spectral wavelength prediction result, and control the lighting in the breeding area for illumination according to the lighting control plan.

[0060] Specifically, according to the predicted proportion of spectra of different wavelengths, select a suitable adjustable light source to mix and output the required spectrum. Optional light sources include LEDs, etc. Calculate the area of ​​the target breeding area, and determine the required total light intensity in combination with the spectrum wavelength results. And calculate the specific output intensity of each light source based on the proportion of light of different wavelengths. According to the structure and layout information of the breeding site, design the placement of the light source so that the light reaches the chicken activity area. Calculate and determine the appropriate lighting angle. By determining the type of light source, adjusting the output intensity and calculating the lighting angle, a directly executable lighting plan is generated to perform auxiliary lighting for the breeding area.

[0061] For example, the spectrum wavelength results obtained by the neural network prediction model are: red light accounts for 30%, orange light accounts for 20%, blue light accounts for 25%, and white light accounts for 25%. The generated lighting control scheme includes: selecting a mixable LED light source, and calculating the total light intensity required for the area of ​​​​100 square meters of chicken activity area and the spectrum ratio. Calculate the specific intensity of various lights: red light intensity is 2400 lumens, orange light intensity is 1600 lumens, blue light intensity is 2000 lumens, and white light intensity is 2000 lumens. According to the position of the chickens, the LED lighting angle is calculated to be 60 degrees, and the chicken area is located and illuminated. By determining the above parameters, a directly executable LED lighting control scheme is formed to give play to the role of light in regulating the distribution position of the chickens and the brightness of the ambient light, thereby improving the capture efficiency.

[0062] In another optional embodiment of the present application, in order to improve the efficiency and accuracy of capturing specific chickens, the method also includes a precise lighting step for the target chickens to be captured. First, the system obtains image information of the target chickens to be captured through an image acquisition device. This image information may come from the above-mentioned infrared camera device for obtaining the image of the entire chicken flock, or it may be a high-resolution camera specifically used for the capture task.

[0063] After acquiring the image, the system uses an image recognition algorithm similar to the aforementioned convolutional neural network model to analyze the image information of the target chicken to be captured, thereby determining the specific location information of the target chicken to be captured in the target breeding area. The accuracy of this step directly affects the effect of subsequent lighting, so the system may combine information from multiple cameras or use a deep learning model to improve positioning accuracy.

[0064] After determining the location information of the target chickens to be caught, the system will calculate the optimal lighting range based on this location information. The determination of this lighting range needs to take into account multiple factors, including the size of the target chickens to be caught, the distribution of other chickens around, and the terrain characteristics of the breeding area. The system may use a preset algorithm or lookup table to quickly determine the most suitable lighting range to ensure that the light can fully cover the target chickens to be caught without excessively disturbing other chickens around.

[0065] Based on the calculated illumination range, the system then generates corresponding illumination angle adjustment instructions. These instructions include parameters such as the direction, angle, and brightness of the light, aiming to achieve precise illumination of the target chickens to be caught. The system sends these instructions to the lighting control end of the target breeding area through a preset communication protocol. After receiving the instructions, the lighting control end will adjust the angle and other parameters of the light accordingly to achieve precise illumination of the target chickens to be caught. This lighting process can improve the efficiency of capturing specific chickens. By focusing the illumination on the target chickens to be caught, they can be more visually prominent, making it easier for the capture personnel to quickly locate and capture them. At the same time, because the illumination range is carefully calculated, it can minimize interference with other chickens and avoid causing panic and commotion in the entire flock. This not only improves the capture efficiency, but also reduces the impact on the overall breeding environment, which is conducive to maintaining the overall stability and health of the flock.

[0066] In another optional embodiment of the present application, in order to further improve the capture efficiency and reduce the interference to the target chicken to be captured and the surrounding chickens, the method also includes the step of intelligently analyzing the state of the target chicken to be captured and generating capture suggestions. This step is based on the aforementioned acquisition of the image information of the target chicken to be captured, and is intended to provide more accurate and valuable auxiliary information for the chicken catchers.

[0067] Based on the acquired image information of the target chickens to be captured, advanced computer vision algorithms are used to extract the posture and behavior characteristics of the target chickens to be captured. Posture characteristics may include the chicken's standing posture, head position, wing status, etc., while behavioral characteristics may include movement speed, activity frequency, interaction with other chickens, etc. The extraction of these features utilizes deep learning models such as convolutional neural networks or recurrent neural networks, which are trained with a large amount of historical data and can accurately identify and classify various postures and behaviors of chickens.

[0068] Next, the system inputs the extracted posture features and behavior features into a preset state feature mapping table. This mapping table is based on a lot of farming experience and animal behavior research. It associates various posture and behavior combinations with different resting states of chickens. By searching and matching in this mapping table, the system can determine the current resting state of the target chicken to be captured, such as whether it is in deep rest, shallow rest or alert state.

[0069] The system then combines the resting state information of the target chicken with the previously determined location information to generate comprehensive prompt information. These prompt information not only includes the specific location of the chicken, but also its current resting state and capture suggestions generated based on this information. For example, if the target chicken is in a deep resting state, the system may suggest that the chicken catcher take a more cautious and slow approach; if the chicken is in an alert state, the system may suggest waiting for it to relax its vigilance or choosing another time to capture it.

[0070] Finally, the system sends these comprehensive prompt information to the mobile terminal devices of the chicken catchers through the preset communication protocol. These devices may be smart phones, tablets or special portable devices, which can clearly display the location, status and capture suggestions of the target chickens to be caught, and assist the chicken catchers in catching the target chickens through lighting control.

[0071] The following is a system embodiment of the present application, which can be used to execute the method embodiment of the present application. For details not disclosed in the platform embodiment of the present application, refer to the method embodiment of the present application.

[0072] Reference Figure 3 , is a system of a lighting method based on spectral wavelength provided in an embodiment of the present application, the system comprises: an information acquisition module 301, an image recognition module 302, a model prediction module 303, and a lighting control module 304, wherein: An information acquisition module 301 is used to acquire images of chicken flocks in a target breeding area; An image recognition module 302 is used to determine the chicken distribution position and ambient light brightness in the chicken image according to an image recognition algorithm; The model prediction module 303 is used to input the chicken distribution position and the ambient light brightness into a preset spectrum wavelength prediction model to obtain a spectrum wavelength prediction result; The light control module 304 is used to generate a light control plan according to the spectrum wavelength prediction result, and control the lighting in the breeding area according to the light control plan.

[0073] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0074] The present application also discloses an electronic device. Figure 4 , Figure 4 The electronic device 400 may include: at least one processor 401 , at least one network interface 404 , a user interface 403 , a memory 402 , and at least one communication bus 405 .

[0075] The communication bus 405 is used to realize the connection and communication between these components.

[0076] The user interface 403 may include a display interface and a camera interface. Optionally, the user interface 403 may also include a standard wired interface and a wireless interface.

[0077] The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0078] Among them, the processor 401 may include one or more processing cores. The processor 401 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 402, and calling data stored in the memory 402. Optionally, the processor 401 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 401 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface diagrams and applications, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 401, and it can be implemented separately through a chip.

[0079] Among them, the memory 402 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 402 includes a non-transitory computer-readable storage medium. The memory 402 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 402 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 402 may optionally be at least one storage device located away from the aforementioned processor 401. Refer to Figure 4 , the memory 402 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program of a lighting method based on spectral wavelength.

[0080] exist Figure 4In the electronic device 400 shown, the user interface 403 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 401 can be used to call the application program stored in the memory 402 for a lighting method based on spectral wavelengths. When executed by one or more processors 401, the electronic device 400 executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0081] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0082] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0083] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0085] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0086] The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0087] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A lighting method based on spectral wavelength, characterized in that: include: Acquire images of chicken flocks in the target breeding area; Determine the distribution position of the chickens and the ambient light brightness in the chicken image according to the image recognition algorithm; Input the chicken distribution position and the ambient light brightness into a preset spectral wavelength prediction model to obtain a spectral wavelength prediction result; A lighting control scheme is generated according to the spectral wavelength prediction result, and the lighting in the breeding area is controlled according to the lighting control scheme for illumination.

2. The spectral wavelength-based lighting method according to claim 1, characterized in that: The step of obtaining the image of the chicken flock in the target breeding area includes: Acquire an initial image of the chicken flock in the target breeding area by using an infrared camera; According to a preset collection frequency, the temperature information of the target breeding area is obtained through a temperature sensor; According to a preset coding format, the initial chicken flock image and the temperature information are converted into a corresponding chicken flock image of the target breeding area.

3. The spectral wavelength based lighting method according to claim 1, characterized in that: The image recognition algorithm is a convolutional neural network model, which is obtained by cyclic training of images of chicken flocks in historical breeding areas, image features of chicken flock positions, and ambient light image features.

4. The spectral wavelength-based lighting method according to claim 3, characterized in that: Determining the chicken distribution position and ambient light brightness in the chicken image according to the image recognition algorithm includes: Extracting position features and illumination features in the image of the chicken flock in the target breeding area through the convolutional neural network model; Determine the distribution position of the chicken flock according to the edge pixel coordinates in the position feature; The ambient light brightness is determined according to the pixel grayscale value in the illumination feature.

5. The spectral wavelength based lighting method according to claim 1, characterized in that: Before acquiring the chicken flock image in the target breeding area, the method further includes: Obtaining historical chicken flock distribution positions and historical ambient light brightness, and using the historical chicken flock distribution positions and the historical ambient light brightness as historical evaluation indicators; Determine the historical climate information corresponding to the historical evaluation index, the historical climate information including cloudy days, sunny days and thunderstorm days; Determining an actual spectral wavelength according to the historical evaluation index and the historical climate information corresponding to the historical evaluation index; Inputting the historical evaluation index, the historical climate information corresponding to the historical evaluation index, and the actual spectral wavelength into an initial neural network model to obtain a target spectral wavelength prediction result; The initial neural network model is iteratively trained according to the target spectrum wavelength prediction result to obtain the preset spectrum wavelength prediction model.

6. The spectral wavelength-based lighting method according to claim 5, characterized in that: The determining of the actual spectral wavelength according to the historical evaluation index and the historical climate information corresponding to the historical evaluation index, wherein the spectral wavelength includes blue light, white light, red light and orange light, comprises: Performing weighted summation on the historical chicken flock distribution positions and the historical ambient light brightness in the historical evaluation index to obtain the chicken flock distribution positions and the ambient light brightness evaluation values; If the chicken flock distribution position and the ambient light brightness evaluation value are greater than or equal to the state evaluation value threshold, then increasing the proportion of red light and orange light in the spectral wavelength; If the chicken distribution position and the ambient light brightness evaluation value are less than the state evaluation value threshold, then increasing the proportion of blue light and white light in the spectral wavelength; According to the historical climate information mapping table, determine the proportions of the blue light, the white light, the red light and the orange light in the spectral wavelength corresponding to cloudy days, sunny days and thunderstorm days; The actual spectrum wavelength is obtained according to the corresponding proportions of the blue light, white light, red light and orange light in the spectrum wavelength.

7. The spectral wavelength based lighting method according to claim 1, characterized in that: After generating a lighting control scheme according to the spectral wavelength prediction result and controlling the lighting of the breeding area for illumination according to the lighting control scheme, the method further includes: Obtain image information of target chicken to be captured; Determining the location information of the target chicken to be captured in the target breeding area according to the image information of the target chicken to be captured; Determining the illumination range of the target chicken arrest according to the position information; Based on the illumination range, a lighting angle adjustment instruction is generated and sent to the lighting control terminal of the target breeding area to adjust the lighting angle to illuminate the target arrested chickens.

8. The spectral wavelength-based lighting method according to claim 7, characterized in that: After determining the location information of the target chicken to be captured in the target breeding area according to the image information of the target chicken to be captured, the method further includes: Based on the target arrested chicken image information, extracting the posture features and behavior features of the target arrested chicken; Determine the current resting state of the target arrested chicken in a preset state feature mapping table according to the posture feature and the behavior feature; In combination with the current resting state and position information of the target captured chicken, prompt information is generated and sent to the chicken-catching personnel terminal.

9. A lighting method system based on spectral wavelength, characterized in that: The system comprises: An information acquisition module (301) is used to acquire images of chicken flocks in a target breeding area; An image recognition module (302) is used to determine the distribution position of the chickens and the brightness of the ambient light in the chicken image according to an image recognition algorithm; A model prediction module (303) is used to input the chicken distribution position and the ambient light brightness into a preset spectrum wavelength prediction model to obtain a spectrum wavelength prediction result; The lighting control module (304) is used to generate a lighting control plan according to the spectral wavelength prediction result, and control the lighting of the breeding area according to the lighting control plan.

10. An electronic device, characterized in that: The electronic device (400) comprises a processor (401), a memory (402), a user interface (403) and a network interface (404), wherein the memory (402) is used to store instructions, the user interface (403) and the network interface (404) are used to communicate with other devices, and the processor (401) is used to execute the instructions stored in the memory (402) so that the electronic device (400) executes the spectral wavelength-based lighting method according to any one of claims 1 to 8.

Images