KR102250240B1 - Smart barn management system with LED sterilization and management method thereof - Google Patents

Abstract

The present invention relates to a smart LED sterilizing livestock barn management system which predicts the occurrence of a disease through a pre-learned deep learning algorithm and, as a follow-up measure, automatically prevents the occurrence of a disease through LED sterilization when a disease occurrence probability derived from deep learning exceeds a specific threshold, and to a management method thereof. The smart LED sterilizing livestock barn management system according to the present invention comprises: a sensor unit including an indicator virus detection sensor, an indicator bacteria detection sensor, a temperature sensor, and a humidity sensor; a pre-learning unit including a database for storing biological indicator data and environmental indicator data of a livestock burn, and a disease learning unit for outputting a completed learning model after training a deep neural network through the data input from the database; an integrated disease management system including a data collection unit which receives and normalizes the biological indicator data and the environmental indicator data detected by the sensor unit, a disease prediction unit which receives the deep neural network from the disease learning unit, receives the data of the data collection unit, calculates a predicted value for each disease, and converts the predicted value into a reliability score, and a lighting control unit which receives the reliability score; and a lighting device including an LED lighting unit which outputs the LED light of a specific wavelength capable of performing sterilization, and configured to emit the LED light of the specific wavelength according to a signal transmitted from the integrated disease management system.

Description

LED sterilization smart barn management system and its management method {Smart barn management system with LED sterilization and management method thereof}

The present invention relates to an LED sterilization smart livestock management system through prediction of livestock diseases, and more specifically, derived through the collection of indicator viruses and indicator bacteria for viral and bacterial diseases that may occur in industrial animals of individual livestock houses. Predicting disease occurrence through deep learning algorithms learned in advance based on existing databases using biological indicators and environmental indicators including temperature and humidity, and as a follow-up measure, the probability of disease occurrence derived from deep learning is at a certain threshold. It relates to an LED sterilization smart livestock house management system that automatically prevents diseases through LED sterilization when the value is exceeded and its management method.

In the past, research on a method of predicting the occurrence of diseases and performing preventive sterilization has been steadily conducted in order to prevent viral diseases that constantly cause large-scale livestock mortality. In particular, due to the development of sensors and the advent of artificial intelligence, various attempts have been made with the aim of implementing a smart house that can automatically perform prediction and sterilization.

As an example of such a smart congratulatory implementation, Korean Patent Publication No. 10-2020-0063648 (published on Jun. 5, 2020) discloses a management system that monitors diseases through information on images, temperature, gas, and saliva. The system predicts the occurrence of a disease by analyzing hydrodynamic effects based on the information and provides the result in virtual/augmented reality.

Korean Patent Publication No. 10-2020-0059140 (published on May 28, 2020) discloses a management system that monitors disease by extracting the body temperature of an object of a corresponding industrial animal through a thermal imaging camera. It supports monitoring by object by attaching a sensor.

Republic of Korea Patent Registration No. 10-2102732 (announced on April 21, 2020) is a livestock house management system that sprays sterilizing liquid into the livestock in a mist form using a plurality of spray nozzles when certain conditions are met after measuring the temperature and humidity in the livestock. To this end, a separate chemical tank must be installed and an additional control unit for ventilation in the livestock house is required.

Republic of Korea Patent Registration No. 10-1624929 (Announcement 20160530) includes an LED lighting part that irradiates the LED light, a heating part that surrounds the lighting part and generates heat toward the front, and positive and negative ions formed inside the lighting part. An illuminator consisting of an ion generator that simultaneously generates and purifies the surrounding air, and a control unit that adjusts and controls the heating temperature of the heating part through wireless communication along with the amount of light and the intensity of the lighting of the lighting part in consideration of a regional average temperature. An LED lighting system for growing light comprising a is disclosed.

However, in the case of the conventional smart livestock management systems as described above, only one technology of monitoring the disease or sterilizing the livestock houses is implemented, so that the occurrence of diseases is predicted through an algorithm and at the same time, the livestock houses can be automatically sterilized as follow-up measures The function cannot be achieved, and there is a problem in that the livestock farmers who are in actual use cannot predict diseases of industrial animals and must adopt several systems.

In addition, the conventional smart livestock farm management systems as described above have a problem in that it imposes a financial burden on livestock farms by requiring the installation of an additional driving unit in the livestock house for the implementation of the technology.

In addition, the conventional smart livestock management systems as described above have a high possibility of negatively affecting the corresponding industrial animals because the method of collecting or sterilizing information is invasive, and there is a problem that is inefficient and impractical to realize due to the need for periodic replacement.

In addition, in the case of the conventional lighting device as described above, not only is it easy to propagate harmful bacteria in the air, but it is possible to provide only the lighting function by the LED element despite the environment in which various odor-causing factors that cause odors are prone to occur. Accordingly, there was a problem that it was difficult to improve the livestock environment.

1.Korean Patent Publication No. 10-2020-0063648 (published on Jun. 5, 2020) 2. Korean Patent Publication No. 10-2020-0059140 (published on May 28, 2020) 3. Korean Patent Registration No. 10-2102732 (Announcement on April 21, 2020) 4. Korean Patent Registration No. 10-1624929 (2016.05.30 announcement)

The present invention was conceived to solve the problems of the prior art, and an object of the present invention is to predict the occurrence of diseases through biological indicator data such as indicator virus and indicator bacteria and environmental indicator data such as temperature and humidity. Algorithms can be effectively customized for each livestock species, and after installing the indicator microbial detection sensor according to the livestock disease, the measured real-time biological indicators of the indicator virus and indicator bacteria and the environmental indicator data measured by the temperature/humidity sensor provided in the livestock house are stored. Through this, it is possible to immediately predict the occurrence of livestock diseases, and to provide an LED sterilization smart livestock management system and a management method that can efficiently control the LED for sterilization according to the prediction of disease occurrence.

In addition, another object of the present invention is to provide a sterilization function in addition to the original lighting function as the sterilizing LEDs are arranged around the lighting LEDs of the lighting device, and tourmaline is mixed when the radiator of the lighting device and the lamp shade are manufactured. LED sterilization smart livestock house management system with a sterilization function in which the deodorization function can be provided due to negative ions and far-infrared rays by injection molding, and the lighting device is moved to the required position along the moving guide frame installed on the upper part of the house And it is to provide a management method.

In order to achieve the above object, the LED sterilization smart livestock management system according to the present invention includes: a sensor unit including an indicator virus detection sensor, an indicator bacteria detection sensor, a temperature sensor, and a humidity sensor; A pre-learning unit including a database for storing biological indicator data and environmental indicator data of a livestock house, and a disease learning unit for outputting a completed learning model after learning a deep neural network through data input from the database; A data collection unit that receives and normalizes the biological indicator data and environmental indicator data detected by the sensor unit, and receives the deep neural network from the disease learning unit and receives data from the data collection unit to calculate a predicted value for each disease. A disease management integrated system including a disease prediction unit that converts this into a confidence score, and a lighting control unit receiving the confidence score; And an LED lighting unit that outputs an LED light of a specific wavelength capable of sterilization, and irradiates the LED light of the specific wavelength according to a signal transmitted from the disease management integrated system. It characterized in that it comprises a.

In addition, in the LED sterilization smart livestock management system according to the present invention, the disease learning unit, a data input unit for inputting the data of the database; A model learning unit that completes a learning model by learning a deep neural network through the received data; And a learning model output unit that outputs the completed learning model to the disease management integrated system. It characterized in that it comprises a.

In addition, in the LED sterilization smart livestock management system according to the present invention, the deep neural network is learned to output a confidence score by calculating the probability of occurrence for each of the representative diseases according to the breeding livestock based on the input data.

In addition, in the LED sterilization smart livestock management system according to the present invention, the disease prediction unit comprises: a data input unit for inputting biological and environmental index data; A prediction model unit that receives the deep neural network from the pre-learning unit and calculates a predicted value for representative diseases according to breeding breeds based on data input from the data input unit; A prediction value output unit outputting a prediction value of the prediction model unit; And a confidence score output unit converting the predicted value into a confidence score for each disease and transmitting the confidence score to the lighting control unit.

In addition, in the LED sterilization smart house management system according to the present invention, the lighting control unit checks whether the reliability of a specific disease exceeds a threshold value based on the confidence score and disease type data transmitted from the disease prediction unit to signal A central control device that transmits the signal; And a lighting control device that transmits a lighting signal of a specific wavelength value for sterilizing the disease pathogen according to a signal transmitted from the central control device to a wireless communication module attached to the lighting device through a gateway. .

In addition, in the LED sterilization smart livestock management system according to the present invention, the specific wavelength of the lighting device is characterized in that 405nm.

In addition, in the LED sterilization smart barn management system according to the present invention, the lighting device includes: an LED lighting unit provided with an LED for illumination and an LED for sterilization on a lower surface of a substrate and a lamp shade at the edge of the substrate; A radiator provided above the substrate of the LED lighting unit to emit heat generated from the substrate to the outside; And a power supply provided on the upper part of the radiator, connected to the radiator in a state spaced apart from the upper end of the radiator under the interposition of a spaced connection member, and receiving power through a power line to emit light of the lighting LED and the sterilization LED. It characterized in that it includes.

In addition, in the LED sterilization smart barn management system according to the present invention, the lampshade and the radiator of the LED lighting unit are manufactured by injection molding a synthetic resin material mixed with tourmaline.

In addition, in the LED sterilization smart livestock house management system according to the present invention, the upper end of the power supply unit is characterized in that it is provided with a moving fixture that is moved in position along the moving guide frame installed on the upper part of the livestock house.

In addition, in the LED sterilization smart barn management system according to the present invention, a mounting ring is provided at the upper end of the power supply unit, and the mounting ring is supported by being hooked on a moving mounting hook that is moved in position along a moving guide frame installed at the top of the barn house. It is characterized by being.

In addition, the LED sterilization smart livestock management method according to the present invention includes a sensing step of collecting biological index data and environmental index data in the livestock house; A disease that completes a learning model after learning a deep neural network to output a confidence score by calculating the probability of occurrence for each of the representative diseases according to breeding breeds based on the data obtained by normalizing the biological and environmental indicator data. Learning stage; Disease prediction in which a predicted value for each representative disease is calculated by receiving the deep neural network and inputting the biological and environmental indicator data acquired in the sensing step into the deep neural network, and converting the calculated predicted value into a confidence score for each disease step; And it checks whether the reliability of a specific disease exceeds the threshold value by the confidence score and the disease type data, and when it is exceeded, the LED lighting signal of a specific wavelength value for sterilizing the pathogen of the disease is transmitted to the LED light. It includes; lighting control and sterilization step to sterilize by.

The LED sterilization livestock management system, which can efficiently control the sterilization LED according to the disease outbreak prediction of the present invention having the above configuration, provides biological indicators such as indicator viruses and indicator bacteria in the housing, and environmental conditions such as temperature and humidity of the housing. Forecasts through indicator data are made in real time, and continuous sterilization is possible regardless of the working hours of livestock farmers. Therefore, it is possible to prevent large-scale infectious diseases in advance by automatic LED sterilization through disease prediction, thereby preventing large-scale livestock death.

In addition, according to the present invention, there is an advantage of minimizing stress from the LED received by livestock through automatic control of the sterilization LED according to prediction using biological and environmental indicator data.

In addition, according to the present invention, there is an advantage in that the cost of preventing livestock diseases can be reduced through the LED sterilization smart livestock management system as described above, and the cost of recovering damage from livestock diseases caused by viruses and bacteria is minimized. .

In addition, according to the present invention, there is an advantage of being able to grasp the state of virus and bacterial livestock diseases in the country through real-time collection of biological and environmental indicator data.

In addition, according to the present invention, since deep learning is performed using biological indicators of representative diseases for each livestock type and environmental indicators in the livestock house, it is possible to predict diseases tailored to each livestock type more efficiently than a general disease diagnosis system, and the LED sterilization system efficiently through prediction information. You will be able to control it.

In addition, according to the present invention, as the LED for sterilization is arranged around the LED for lighting of the lighting device equipped with the sterilization function of the LED sterilization smart livestock house management system, the original lighting function is provided by the LED for lighting and at the same time, harmful by the LED for sterilization. There is an advantage that can provide even sterilization function for bacteria.

In addition, according to the present invention, as the radiator of the lighting device equipped with the sterilization function of the LED sterilization smart house management system and the lamp shade are injection-molded in a mixed state of tourmaline, the deodorizing function due to negative ions and far-infrared rays generated from tourmaline is Can be provided.

In addition, according to the present invention, there is an advantage that the lighting device having the sterilization function of the LED sterilization smart livestock house management system can be moved to a required position when necessary along the grid-shaped movement guide frame installed on the upper part of the livestock house and used.

1 is a structural diagram showing an LED sterilization smart livestock house management system according to an embodiment of the present invention.
Figure 2 is a block diagram showing the configuration of the LED sterilization smart livestock house management system according to an embodiment of the present invention.
Figure 3 is a block diagram showing a disease prediction unit of the LED sterilization smart livestock house management system according to an embodiment of the present invention.
Figure 4 is a block diagram showing a disease learning unit of the LED sterilization smart livestock house management system according to an embodiment of the present invention.
Figure 5 is a perspective view of a lighting device having a sterilization function of the LED sterilization smart livestock house management system according to an embodiment of the present invention.
Figure 6 is a bottom perspective view of a lighting device having a sterilization function of the LED sterilization smart livestock house management system according to an embodiment of the present invention.
7 is a conceptual diagram illustrating a sterilization process of a lighting device having a sterilization function of the LED sterilization smart livestock house management system according to an embodiment of the present invention.
8 is a graph showing porphyrin absorbance by wavelength in the reaction process of FIG. 7.
9 is a graph showing a sterilization test result of a lighting device having a sterilization function of the LED sterilization smart livestock house management system according to an embodiment of the present invention.
10 is a photograph showing a sterilization test result of a lighting device having a sterilization function of the LED sterilization smart livestock house management system according to an embodiment of the present invention.
11 is a perspective view of a lighting device having a sterilization function of the LED sterilization smart livestock house management system according to another embodiment of the present invention.
12 is a use state diagram of a lighting device having a sterilization function of the LED sterilization smart livestock house management system according to another embodiment of the present invention.
13 is a perspective view of a lighting device having a sterilization function of the LED sterilization smart livestock house management system according to another embodiment of the present invention.
Figure 14 is a flow chart of the LED sterilization smart livestock management method according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. This is for explaining in detail enough that one of ordinary skill in the art can easily carry out the invention, and thus the technical spirit and scope of the present invention are not limited thereto.

In describing the embodiments of the present invention, descriptions of technical contents that are well known in the technical field to which the present invention pertains and are not directly related to the present invention will be omitted. This is to more clearly convey the gist of the present invention by omitting unnecessary description. For the same reason, some components in the accompanying drawings are exaggerated, omitted, or schematically illustrated.

2 is a block diagram showing the configuration of an LED sterilization smart livestock house management system according to an embodiment of the present invention.

As shown in Fig. 2, the LED sterilization smart house management system according to an embodiment of the present invention includes a sensor unit 29, a learning unit 28, an integrated disease management system 21, and a lighting device 10. do.

The sensor unit 29 is composed of a temperature sensor 20, a humidity sensor 19 for collecting environmental indicators, an indicator virus detection sensor 18 for collecting biological indicators, and an indicator bacteria detection sensor 17. Since the temperature sensor 20 and the humidity sensor 19 utilize a sensor that is basically built into the individual house 16, it does not require additional installation in the house 16, but the indicator virus detection sensor 18 and the indicator bacteria detection sensor ( 17) can be additionally installed according to the types of viruses and bacteria that are indicators for predicting other representative diseases according to the breeding breed of the house (16).

Indicator viruses and indicator bacteria collectively refer to microorganisms that serve as indicators that indirectly inform the occurrence of pathogenic microorganisms. Since pathogenic microorganisms are of various types, it is impractical to individually test them. Therefore, tests using indicator microorganisms that are easy to detect and distinguish with high possibility of coexistence with pathogenic microorganisms have been widely used in existing environmental evaluations.

Unlike sensors for measuring environmental indicators, in order for a sensor for detecting indicator microorganisms to measure the concentration of microorganisms in the air, a process in which the microorganisms must be drawn to an area where the sensor can be measured is required. This is called capture.

The types of indicator microorganism detection sensors can be largely divided according to this collection method, and the suitability of each collection method varies depending on the microorganism targeted for detection.

Types of sensors according to the collection method include a mass-sensitive-based measurement sensor, an impedance-based measurement sensor, a current and voltage-based measurement sensor, an optical fiber-based measurement sensor, and a surface plasmon resonance-based measurement sensor. It does not mean limiting.

Next, the learning unit 28 will be described.

As shown in FIG. 2, the learning unit 28 learns a deep neural network through a database 27 storing biological indicator data and environmental indicator data of a livestock house, and data input from the database 27. It includes a disease learning unit 26 for outputting the completed learning model.

Figure 4 is a block diagram showing a disease learning unit of the LED sterilization smart livestock management system according to an embodiment of the present invention, the disease learning unit 26 of the pre-learning unit 28 is a first data input unit 260, deep It is composed of a model learning unit 261 configured with a running deep neural network and a learning model output unit 262.

The disease learning unit 26 classifies the existing biological and chemical indicator data into a database by classifying the existing biological and chemical indicator data by virus type, virus detection concentration, temperature, humidity, and disease occurrence, and performs pre-processing through normalization before inputting into the deep learning deep neural network. Then, the preprocessed data is learned through a deep learning deep neural network.

The first data input unit 260 receives biological index data and environmental index data from the livestock environment database 27 previously secured, performs normalization, and then inputs the data to the model learning unit 261. The model learning unit 261 trains the deep neural network through the received data and then outputs the completed learning model through the learning model output unit 262. The deep neural network is learned to output a confidence score by calculating the probability of occurrence for each of the representative diseases according to breeding livestock based on the input livestock environment data.

Next, the integrated disease management system 21 will be described.

As shown in FIG. 2, the integrated disease management system 21 collects and normalizes biological indicator data such as indicator viruses and indicator bacteria detected by the sensor unit 29, and environmental indicator data such as temperature and humidity. Disease prediction that receives the deep neural network from the data collection unit 24 and the disease learning unit 26, receives data from the data collection unit 24, calculates a predicted value for each disease, and converts it into a confidence score. It includes a unit 23 and a lighting control unit 22 for controlling the sterilization LED when the disease prediction confidence score derived by receiving the confidence score exceeds a specific threshold value.

The data collection unit 24 collects environmental indicator data such as temperature and humidity measured by sensors provided in the livestock house, and installs a sensor capable of detecting indicator viruses and indicator bacteria used to predict specific diseases. After measurement, the data is collected and transmitted to the disease prediction department.

The disease prediction unit 23 derives a disease prediction value by pre-processing the data collected by the data collection unit 24 to the deep neural network learned by the disease learning unit 26 through deep learning.

Figure 3 is a block diagram showing a disease prediction unit of the LED sterilization smart livestock management system according to an embodiment of the present invention, the disease prediction unit 23 is derived from the second data input unit 230, the disease learning unit 26 It is composed of a learning model unit 231, a prediction model unit 232 composed of a deep learning deep neural network, a prediction value output unit 233, and a confidence score output unit 234.

The second data input unit 230 receives the data secured through the sensors in the livestock house 16, performs normalization, and then inputs the data to the prediction model unit 232. The prediction model unit 232 receives the deep neural network of the disease learning unit 26 through the learning model output unit 262 and inputs data received through the second data input unit 230 to obtain a predicted value for each representative disease. Calculate. The calculated predicted value is converted into a confidence score for each disease in the confidence score output unit 234, and the corresponding confidence score is transmitted to the lighting control unit 22, which will be described later, through the confidence score output unit 234.

The lighting device 10 includes an LED lighting unit 110 that outputs a specific wavelength capable of sterilization, and a wireless communication module 15 capable of receiving a signal generated from the gateway 14 of the integrated disease management system 21 is provided. Attached.

Next, representative diseases for each species that are the target of prediction in the learning model are described. However, this does not mean the limitation of the technical spirit and scope of the present invention.

According to the data on the outbreak of livestock infectious diseases that occurred from 2017 to 2020 from the Ministry of Agriculture, Forestry and Livestock Quarantine, tuberculosis and Brucella disease in cattle, swine genital respiratory syndrome, African swine fever, and poultry typhus and Chubaekri in chicken It is a representative infectious disease that causes great economic losses to people.

Tuberculosis is classified as a domestic legally classified as a second-class livestock infectious disease, and is infected by Mycobacterium bovis and is a common contagious disease. In addition, inhalation of Mycobacterium tuberculosis through the respiratory tract is known as the most common method. Depending on environmental conditions, it can survive up to 14 days on land and up to 3 months in water. Brucella disease is a disease that is mainly infected by cattle by Brucella abortus, and is a disease that causes maternity and premature birth in the second half of pregnancy. It is commonly transmitted through direct contact with the mucous membranes of infected individuals, and it is known that Brucella bacteria can survive from a minimum of one day to a maximum of 8 months on the ground depending on conditions.

Porcine reproductive and respiratory syndrome virus, which causes the largest economic loss as a single disease in pigs in Korea, is infected by reproductive disorders such as late pregnancy abortion or preterm birth in breeding pigs, and respiratory symptoms in piglets. It is a disease that causes a decrease in weight gain and an increase in mortality. Transmission occurs primarily through direct or indirect contact with the body fluids of an infected individual or through the respiratory tract. It is known to survive from a minimum of 3 days to a maximum of 60 days depending on environmental conditions. African swine fever is infected by the African swine fever virus. The morbidity rate is high, and the mortality rate is almost 100% when infected with an acute type, so it is a disease that causes tremendous damage to the pig farming industry. It is mainly transmitted through direct contact with the body fluids of an infected individual, and the greatest characteristic is that it can also be transmitted through contaminated objects. It is known to survive in contaminated media for a minimum of 3 days to a maximum of 18 months depending on environmental conditions.

Poultry tea booth and chubaekri, one of the representative infectious diseases of chickens, are caused by Salmonella gallinarum and Salmonella pullorum, respectively, and are acute and chronic infectious diseases that cause sepsis, causing white diarrhea in chicks. It is a disease that does not show symptoms in the sexual system, causes death, lowers the spawning rate, and causes diarrhea. In particular, this disease is a disease that requires attention because it can be transmitted through the egg zone. Pathogens are known to survive up to 20 days in distilled water and water.

Next, the theory of the deep neural network and the algorithm of the present invention for predicting the disease will be described in detail.

Prior to the description of the disease prediction algorithm, a brief introduction is made to a learning method of deep learning, which is one of the types of artificial intelligence to be used in the present invention. Deep learning is defined as a machine learning algorithm that attempts a high level of abstraction through a combination of several nonlinear transducers, and is a field of machine learning that teaches computers how to think of humans in a large framework. When there is any data, a lot of research is being conducted to express it in a form that can be recognized by a computer and apply it to learning. As a result of these efforts, techniques such as neural networks through deep learning have been applied to disease prediction, leading to state-of-the-art Show the results. Deep neural networks have the advantage of being able to process desired multi-classification tasks with good performance by expressing them as column vectors.

A feature of the present invention is that the AI learning result does not affect only disease classification, but the learning for disease prediction by functionalizing and estimating conditions in which a disease can occur, and disease prediction and automatic control of LED sterilization through the use thereof. Point.

As shown in FIG. 4, the disease learning unit 26, which derives an existing model for predicting a disease from the previous learning unit 28, includes a first data input unit 260, a model learning unit 261, and a learning model output unit ( 262).

The first data input unit 260 classifies the biological and chemical indicator data in the database 27 into virus/bacteria type, virus/bacteria detection concentration, temperature, humidity, and disease occurrence, and performs normalization for input into the deep neural network. And performs a role of inputting into the model learning unit 261. The model learning unit 261 trains the deep neural network through the received data and then outputs the completed learning model through the learning model output unit 262. The deep neural network is learned to output a confidence score by calculating the probability of occurrence for each of the representative diseases according to breeding livestock based on the input livestock environment data.

The data collection unit 24 of the integrated disease management system 21 receives biological indicators such as indicator viruses and indicator bacteria detected by the sensor unit 29 and environmental indicator data such as temperature and humidity, and receives the disease prediction unit 23. Normalization process is performed before input into.

As shown in FIG. 3, the disease prediction unit 23 includes a second data input unit 230, a learning model unit 231, a prediction model unit 232, a prediction value output unit 233, and a confidence score output unit 234. It consists of

The second data input unit 230 performs normalization in order to input biological and chemical indicator data, that is, virus/bacteria type, virus/bacteria detection concentration, temperature, humidity, and disease occurrence into the deep neural network, and then a prediction model unit ( 232). The prediction model unit 232 receives the above-described deep neural network from the learning model output unit 262 as an existing model and inputs data through the second data input unit 230 to calculate a predicted value for each representative disease. The predicted value output unit 233 outputs a predicted value derived through the predictive model unit 232. The calculated predicted value is converted into a confidence score for each disease in the predicted value output unit 233 and transmitted to the lighting control unit 22 through the corresponding confidence score confidence score output unit 234. That is, the disease prediction unit 23 outputs the accompanying confidence score when predicting each disease along with the disease type and transmits the output to the lighting control unit 22.

Next, the lighting control unit 22 of the present invention will be described in detail.

1 is a structural diagram showing an LED sterilization smart livestock house management system according to an embodiment of the present invention, the lighting control unit 22 is composed of a central control device 12, a lighting control device 13, a gateway 14 .

The central control unit 12 checks whether the reliability of any one disease exceeds a specific threshold value based on the confidence score and disease type data transmitted from the disease prediction unit 23 to the lighting control unit 22. When such data is confirmed, the central control device 12 transmits a signal to the lighting control device 13. The lighting control device 13 transmits a lighting signal having a specific wavelength value for sterilizing the pathogen of the disease to the wireless communication module 15 attached to the LED lighting device 10 through the gateway 14. The LED lighting device 10 continues the lighting gradually increasing to a certain brightness according to the transmitted wavelength value until the turn-off signal is transmitted. When the predicted reliability value falls below the threshold value, a light-off signal that gradually decreases in brightness is transmitted through the same sequence, thereby providing a livestock house management system according to the disease prediction result.

The lighting device 10 includes an LED lighting unit that outputs a specific wavelength capable of sterilization, and a wireless communication module capable of receiving a signal generated from the gateway of the integrated management system is attached.

5 is a perspective view of a lighting device having a sterilization function of the LED sterilization smart house management system according to an embodiment of the present invention, and FIG. 6 is a sterilization function of the LED sterilization smart house management system according to an embodiment of the present invention. It is a bottom perspective view of a lighting device.

As shown, the lighting device 10 having a sterilization function of the LED sterilization smart livestock house management system according to an embodiment of the present invention has a lighting LED 112 provided at the center on the lower surface of the substrate 111, and the At the edge of the substrate 111, the LED lighting unit 110 provided with the lamp shade 113, and the heat generated from the substrate 111 is provided on the upper portion of the substrate 111 of the LED lighting unit 110 to the outside. It is provided on the top of the radiator 120 and the radiator 120, and is connected to the radiator 120 in a state spaced apart from the top of the radiator 120 with the spaced connection member 131 interposed therebetween, and is connected to the radiator 120 through a power line. A power supply unit 130 for emitting power to the lighting LED 112 and the sterilization LED 114 is included, and a wireless communication module 15 is attached.

Hereinafter, the components of the lighting device 10 having a sterilization function of the LED sterilization smart livestock house management system according to an embodiment of the present invention and the relationship between the components will be described in detail with reference to the accompanying drawings. .

As shown in Figs. 5 and 6, the lighting device 10 having the sterilization function of the LED sterilization smart livestock house management system according to an embodiment of the present invention is installed on the upper part of the livestock house to provide illumination light to the livestock house, An LED lighting unit 110, a radiator 120 provided on an upper surface of the substrate 111 of the LED lighting unit 110, and a power supply unit 130 are included.

The LED lighting unit 110 corresponds to a kind of light-emitting lamp that provides illumination light in the barn, and has a structure of an LED luminaire in which a plurality of LEDs for illumination 112 are provided on the lower surface of the substrate 111. The lighting LED 112 may be formed of an LED that emits light having a wavelength band in a visible light region such as white light.

An illumination shade 113 is provided at the edge of the substrate 111. The lamp shade 113 serves to induce the illumination light generated from the lighting LED 12 to be radiated downward, and is provided in a form that extends obliquely to the outside as it goes downward.

The lampshade 13 of the LED lighting unit 110 may be injection-molded only with a synthetic resin material, but it is preferable that a synthetic resin material mixed with tourmaline is injection-molded.

Tourmaline has a hexagonal crystal form and is a crystalline boron silicate mineral in which elements such as aluminum, iron, magnesium, sodium, lithium or potassium are mixed. Electricity is generated by friction. When heated, both ends of the crystal are positive and negative. It is also called tourmaline because it has the property of being charged with poles. These tourmalines emit negative ions and far-infrared rays, and the negative ions and far-infrared rays emitted from tourmaline play a role of exerting a deodorizing function by removing various odor-causing factors that cause odor in the livestock house.

As shown in FIG. 6, the sterilizing LED 114 is arranged on the lower surface of the substrate 111 to surround the lighting LED 112, and the position thereof is not limited thereto and may be configured in various ways.

The sterilizing LED 114 is an LED that emits light having a wavelength range of 405 nm, and serves to create a hygienic and safe livestock environment by sterilizing various harmful bacteria in the livestock house.

7 is a conceptual diagram illustrating a sterilization process of a lighting device having a sterilization function of an LED sterilization smart livestock house management system according to an embodiment of the present invention, as shown, light having a wavelength of 405 nm from the sterilizing LED 114 When irradiated to this bacterium, a reaction with porphyrin occurs within the bacterium, and reactive oxygen species are produced, explaining the process of destroying the bacterial cell.

Porphyrin is a reactant that exists in bacteria and absorbs the wavelength of 405 nm the most.

FIG. 8 is a graph showing the absorbance of porphyrin for each wavelength in the reaction process of FIG. 7, and shows the result of absorbing porphyrin at a maximum value at a wavelength of 405 nm.

Since the 405 nm wavelength is in the visible range of 400 to 800 wavelength, the sterilizing LED 114 of the 405 nm wavelength of the lighting device of the present invention enables sterilization while being harmless to humans and livestock.

Conventional LED sterilizers are harmful to humans and livestock because they sterilize using wavelengths in the ultraviolet range of UV-A (320nm~400nm), UV-B (280nm~320nm), and UV-C (100nm~280nm). , As the present invention is sterilized by the sterilization wavelength in the visible light region, it is not harmful to the human body and livestock, and exhibits an excellent sterilization effect as shown below.

9 is a graph showing the result of a sterilization test of a lighting device equipped with a sterilization function of the LED sterilization smart livestock house management system according to an embodiment of the present invention. When 24 hours elapse after the sterilization LED 114 is irradiated, bacteria It shows that the reduction rate will reach 99.9%.

10 is a photograph showing a sterilization test result of a lighting device equipped with a sterilization function of the LED sterilization smart livestock house management system according to an embodiment of the present invention, the test result of the Korea Living Environment Testing Institute (KCL) sterilization function of the present invention When the lighting device equipped with is irradiated to the sample, it shows that E. coli is almost extinct.

As shown in FIGS. 5 and 6, a radiator 120 is provided on the upper surface of the substrate 111 of the LED lighting unit 110. The radiator 120 radiates heat generated from the substrate 111 to the outside to prevent damage to the substrate 111, the lighting LED 112, the sterilization LED 114, and the power supply unit 130 to be described later by heat. It serves to prevent.

The radiator 120 has a container shape with an open lower portion, and a plurality of radiating fins 121 are radially protruded on the outer circumferential surface to increase the heat dissipation area. In addition, in order to maximize the heat dissipation efficiency, the radiator 20 absorbs heat emitted from the substrate 111 and the heat medium that is phase-changed is filled inside, for example, after the open bottom is sealed with a lower cover (not shown). It is preferable to be vacuum-treated.

In addition, the radiator 120 may be injection-molded with only a synthetic resin material or may be made of a metal material, but, like the lighting shade 113, it is preferable that the radiator 120 is injection-molded with a synthetic resin material mixed with tourmaline. At this time, negative ions and far-infrared rays emitted from tourmaline included in the radiator 120 play a role of exerting a deodorizing function by removing various odor-causing factors that cause odor in the livestock house.

In addition, it is preferable that a ceramic coating layer including a ceramic material such as germanium is laminated on the surface of the radiator 120. As the ceramic coating layer laminated on the surface of the radiator 120 has a much greater thermal conductivity than a conventional synthetic resin material, the radiator 120 not only doubles the heat dissipation effect, but also releases anions additionally, resulting in an odor in the house. It doubles the deodorizing function by additionally removing various odor-causing factors that cause odor.

The power supply unit 130 is connected and fixed to the upper portion of the above-described radiator 120. The power supply unit 30 serves to emit light from the lighting LED 112 and the sterilization LED 114 by receiving power through the power line 132, and converts AC power into DC power using a switching transistor. It can be formed by a well-known power supply module (SMPS: Switched Mode Power Supply) using a switch control method.

As shown in FIG. 6, the power supply unit 130 is spaced apart from the upper end of the radiator 120 under the interposition of a spaced connection member 131 such as an angle-type connection bracket or a gap material, and the upper portion of the radiator 120 It is desirable to be connected to and fixed. In this case, the power supply unit 130 is spaced apart from the radiator 120 that is inevitably heated in the process of receiving heat generated from the substrate 111 of the LED lighting unit 110 and discharging it to the outside. As the connection is fixed to, heat transfer from the radiator 120 is blocked, and thus thermal damage to the power supply unit 130 may be minimized.

In addition, as shown in FIG. 5, a fixing fixture 141 may be provided at an upper end of the power supply unit 130 described above. The fixing fixture 141 is to fix the position of the lighting device 10 having a sterilization function according to an embodiment of the present invention at a specific position in the upper part of the barn, for example, a suspension wire or a fixed hook, etc. It is provided to be able to be coupled to a specific position of the barn in a form.

FIG. 11 is a perspective view of a lighting device having a sterilization function of an LED sterilization smart livestock house management system according to another embodiment of the present invention. As shown, a movement guide installed on the upper part of the livestock house is at the upper end of the power supply unit 130 A moving cradle 142 may be provided that is moved in position along the frame 300. In this case, the lighting device 10 having a sterilization function according to another embodiment of the present invention may be moved to a required position within the barn when necessary and used.

12 is a use state diagram of a lighting device having a sterilization function of the LED sterilization smart livestock house management system according to another embodiment of the present invention, as shown in Figs. 11 and 12, a moving guide frame ( The cross-shaped frame member 305 having a channel-shaped cross-section with an open lower portion is connected and fixed by a straight frame member 307 having a channel-shaped cross-section with an open lower portion to form a grid shape as a whole.

In addition, the upper end of the movable fixture 142 is fitted into the cross-shaped frame member 305 or the straight frame member 307 and is integrally provided with a sliding movable end that is movable through the cross-shaped frame member 305 and the straight frame member 307. .

13 is a perspective view of a lighting device having a sterilization function of an LED sterilization smart livestock house management system according to another embodiment of the present invention. While the fixing ring 143 is integrally provided, the upper end is movably coupled to the movement guide frame 300, and the lower end is provided with a movable mounting hook 144 that is supported by hanging the fixing ring 143. ) Can be combined. In this case, the lighting device 10 having a sterilization function according to another embodiment of the present invention may be moved to a required position within the barn and used if necessary.

Hereinafter, with reference to the accompanying drawings will be described the overall operation of the lighting device having a sterilization function of the LED sterilization smart house management system according to an embodiment of the present invention.

The central control device 12 of the LED sterilization smart livestock house management system according to an embodiment of the present invention is to detect any one disease through the confidence score and disease type data transmitted from the disease prediction unit 23 to the lighting control unit 22. Check if the reliability exceeds a certain threshold. When such data is confirmed, the central control device 12 transmits a signal to the lighting control device 13. The lighting control device 13 transmits a lighting signal having a wavelength of 405 nm for sterilizing pathogens of the disease to the wireless communication module 15 attached to the LED lighting device 10 through the gateway 14. The LED lighting device 10 continues the lighting gradually increasing to a certain brightness according to the transmitted wavelength value until the turn-off signal is transmitted. When the predicted reliability value falls below the threshold value, a light-off signal that gradually decreases in brightness is transmitted through the same sequence, thereby providing a livestock house management system according to the disease prediction result.

In the lighting device 10 having a sterilization function according to a preferred embodiment of the present invention, a sterilization LED 114 is arranged on the lower surface of the substrate 111 to surround the lighting LED 12 and the lighting control device 13 In accordance with the control signal of ), the original lighting function may be provided by the lighting LED 112, and at the same time, the sterilization function for harmful bacteria may be provided by the sterilization LED 114.

In addition, the lighting device 10 having a sterilization function according to a preferred embodiment of the present invention is injection-molded in a state in which tourmaline is mixed during the manufacture of the radiator 120 and the lamp shade 113, so that the negative ions generated from tourmaline and the A deodorizing function may be provided due to far-infrared rays, and as the ceramic coating layer is formed on the surface of the radiator 120, the deodorizing function may be doubled by anions additionally emitted from the ceramic coating layer, and the heat dissipation effect may be doubled due to the ceramic coating layer.

In addition, the lighting device 10 having a sterilization function according to a preferred embodiment of the present invention may be moved to a required position when necessary along the grid-shaped movement guide frame 300 installed on the upper part of the barn and used.

Next, a description will be given of the LED sterilization smart livestock management method of the present invention.

The LED sterilization smart livestock management method according to the present invention learns a livestock disease diagnosis algorithm using a deep learning algorithm using biological indicators such as indicator viruses and indicator bacteria and environmental indicator data such as humidity and temperature of the livestock house, By collecting biological/environmental data installed in the house through the learned deep neural network and controlling the sterilization LED when the confidence score derived by predicting a disease exceeds a specific threshold, viruses and bacteria are sterilized, and when it is lower than a specific threshold, power It is an automatic control method to turn off the power.

14 is a flowchart of an LED sterilization smart house management method according to an embodiment of the present invention.

As shown, the LED sterilization smart house management method of the present invention includes a sensing step (S401), a disease learning step (S403), a disease prediction step (S405), a lighting control and a sterilization step (S407).

The sensing step S401 is a step of detecting an indicator virus and an indicator bacteria for collecting temperature, humidity and biological indicators for collecting environmental indicators.

Indicator viruses and indicator bacteria collectively refer to microorganisms that serve as indicators that indirectly inform the occurrence of pathogenic microorganisms. Since pathogenic microorganisms are of various types, it is impractical to individually test them. Therefore, tests using indicator microorganisms that are easy to detect and distinguish with high possibility of coexistence with pathogenic microorganisms have been widely used in existing environmental evaluations.

In the disease learning step (S403), biological indicator data and environmental indicator data are received from the previously secured livestock environment database 27, normalized, and then input to the model learning unit. The model learning unit 261 trains the deep neural network through the received data and then outputs the completed learning model through the learning model output unit 262. The deep neural network is learned to output a confidence score by calculating the probability of occurrence for each of the representative diseases according to breeding livestock based on the input livestock environment data.

In the disease prediction step (S405), after normalization is performed to input the biological and chemical indicator data acquired in the sensing step, namely, virus/bacteria type, virus/bacteria detection concentration, temperature, humidity, and disease occurrence into the deep neural network. It is input to the prediction model unit 232. The prediction model unit 232 receives the above-described deep neural network from the learning model output unit 262 as an existing model and inputs data through the second data input unit 230 to calculate a predicted value for each representative disease. The predicted value output unit 233 outputs a predicted value derived through the predictive model unit 232. The calculated predicted value is converted into a confidence score for each disease in the predicted value output unit 233 and transmitted to the lighting control unit 22 to be described later through the corresponding confidence score confidence score output unit 234. That is, the disease prediction unit 23 outputs the accompanying confidence score when predicting each disease along with the disease type and transmits the output to the lighting control unit 22.

Next, the lighting control and sterilization steps of the LED sterilization smart house management method of the present invention will be described.

The central control unit 12 of the lighting control unit 22 checks whether the reliability of any one disease exceeds a specific threshold value based on the confidence score and disease type data transmitted in the disease prediction step described above. When such data is confirmed, the central control device 12 transmits a signal to the lighting control device 13. The lighting control device 13 transmits a lighting signal having a specific wavelength value for sterilizing the pathogen of the disease to the wireless communication module 15 attached to the LED lighting device 10 through the gateway 14. The LED lighting device 10 keeps lighting gradually increasing to a certain brightness according to the transmitted 405 nm wavelength value until the light-off signal is transmitted. When the predicted reliability value falls below the threshold value, a light-off signal is transmitted in which the brightness gradually decreases through the same sequence.

The embodiments of the present invention have been exemplarily described above, and may be embodied in various forms without departing from the spirit and scope of the present invention. Accordingly, all embodiments within the scope of the appended claims and their equivalents are included within the scope of the present invention.

Meanwhile, the present specification and drawings disclose preferred embodiments of the present invention, and although specific terms are used, these are merely used in a general meaning to easily explain the technical content of the present invention and to aid understanding of the present invention. It is not intended to limit the scope of the invention.

10 LED lighting
11 Indicator microbial detection sensor
12 Central control unit
13 Lighting control device
14 gateway
15 Wireless communication module
16 congratulatory remarks
17 Indicator bacteria detection sensor
18 Indicator virus detection sensor
19 Humidity sensor
20 temperature sensor
21 Integrated disease management system
22 Lighting control unit
23 Disease Prediction Department
230 Data input section
231 Learning Model Department
232 Prediction model unit
233 Predicted value output section
234 Confidence score output section
24 Data collection unit
26 Disease Learning Department
260 data input unit
261 Model Learning Department
262 Learning model output section
27 databases
28 Skilled Learning Department
29 Sensor part
110 LED Lighting
111 substrate
112 LED for lighting
113 Lampshade
114 LED for sterilization
120 radiator
121 heat dissipation fin
130 Power supply
131 Spacing connection member
132 power line
141 Fixing fixture
142 Moving Jigsaw
143 Retaining Ring
144 Moving Mounting Hook
300 moving guide frame
305 Cross-shaped frame member
307 straight frame member
S401 sensing step
S403 Disease learning stage
S405 disease prediction stage
S407 lighting control and sterilization step

Claims (11)

delete A sensor unit including an indicator virus detection sensor, an indicator bacteria detection sensor, a temperature sensor, and a humidity sensor;
A pre-learning unit including a database for storing biological indicator data and environmental indicator data of a livestock house, and a disease learning unit for outputting a completed learning model after learning a deep neural network through data input from the database;
A data collection unit that receives and normalizes the biological indicator data and environmental indicator data detected by the sensor unit, and receives the deep neural network from the disease learning unit and receives data from the data collection unit to calculate a predicted value for each disease. A disease management integrated system including a disease prediction unit that converts this into a confidence score, and a lighting control unit receiving the confidence score; And
Including; a lighting device for irradiating the LED light of the specific wavelength according to the signal transmitted from the disease management integrated system, including an LED lighting unit that outputs the LED light of a specific wavelength capable of sterilization; and
The disease learning unit,
A data input unit for inputting data of the database;
A model learning unit that completes a learning model by learning a deep neural network through the received data; And
A learning model output unit for outputting the completed learning model to the disease management integrated system;
LED sterilization smart livestock management system comprising a. A sensor unit including an indicator virus detection sensor, an indicator bacteria detection sensor, a temperature sensor, and a humidity sensor;
A pre-learning unit including a database for storing biological indicator data and environmental indicator data of a livestock house, and a disease learning unit for outputting a completed learning model after learning a deep neural network through data input from the database;
A data collection unit that receives and normalizes the biological indicator data and environmental indicator data detected by the sensor unit, and receives the deep neural network from the disease learning unit and receives data from the data collection unit to calculate a predicted value for each disease. A disease management integrated system including a disease prediction unit that converts this into a confidence score, and a lighting control unit receiving the confidence score; And
Including; a lighting device for irradiating the LED light of the specific wavelength according to the signal transmitted from the disease management integrated system, including an LED lighting unit that outputs the LED light of a specific wavelength capable of sterilization; and
The deep neural network is the LED sterilization smart livestock management system, characterized in that the learning to output a confidence score by calculating the probability of occurrence for each of the representative diseases according to the breeding livestock based on the input data. A sensor unit including an indicator virus detection sensor, an indicator bacteria detection sensor, a temperature sensor, and a humidity sensor;
A pre-learning unit including a database for storing biological indicator data and environmental indicator data of a livestock house, and a disease learning unit for outputting a completed learning model after learning a deep neural network through data input from the database;
A data collection unit that receives and normalizes the biological indicator data and environmental indicator data detected by the sensor unit, and receives the deep neural network from the disease learning unit and receives data from the data collection unit to calculate a predicted value for each disease. A disease management integrated system including a disease prediction unit that converts this into a confidence score, and a lighting control unit receiving the confidence score; And
Including; a lighting device for irradiating the LED light of the specific wavelength according to the signal transmitted from the disease management integrated system, including an LED lighting unit that outputs the LED light of a specific wavelength capable of sterilization; and
The disease prediction unit,
A data input unit for inputting biological and environmental indicator data;
A prediction model unit that receives the deep neural network from the learning unit and calculates a predicted value for representative diseases according to breeding breeds based on data input from the data input unit;
A prediction value output unit outputting a prediction value of the prediction model unit; And
And a confidence score output unit for converting the predicted value into a confidence score for each disease and transmitting the confidence score to a lighting control unit. A sensor unit including an indicator virus detection sensor, an indicator bacteria detection sensor, a temperature sensor, and a humidity sensor;
A pre-learning unit including a database for storing biological indicator data and environmental indicator data of a livestock house, and a disease learning unit for outputting a completed learning model after learning a deep neural network through data input from the database;
A data collection unit that receives and normalizes the biological indicator data and environmental indicator data detected by the sensor unit, and receives the deep neural network from the disease learning unit and receives data from the data collection unit to calculate a predicted value for each disease. A disease management integrated system including a disease prediction unit that converts this into a confidence score, and a lighting control unit receiving the confidence score; And
Including; a lighting device for irradiating the LED light of the specific wavelength according to the signal transmitted from the disease management integrated system, including an LED lighting unit that outputs the LED light of a specific wavelength capable of sterilization; and
The lighting control unit checks whether the reliability of a specific disease exceeds a threshold value based on the confidence score and disease type data transmitted from the disease prediction unit, and transmits a signal;
And a lighting control device for transmitting a lighting signal of a specific wavelength value for sterilizing the disease pathogen according to a signal transmitted from the central control device to a wireless communication module attached to the lighting device through a gateway. LED sterilization smart barn management system. The method according to any one of claims 2 to 5,
LED sterilization smart livestock management system, characterized in that the specific wavelength of the lighting device is 405nm. A sensor unit including an indicator virus detection sensor, an indicator bacteria detection sensor, a temperature sensor, and a humidity sensor;
A pre-learning unit including a database for storing biological indicator data and environmental indicator data of a livestock house, and a disease learning unit for outputting a completed learning model after learning a deep neural network through data input from the database;
A data collection unit that receives and normalizes the biological indicator data and environmental indicator data detected by the sensor unit, and receives the deep neural network from the disease learning unit and receives data from the data collection unit to calculate a predicted value for each disease. A disease management integrated system including a disease prediction unit that converts this into a confidence score, and a lighting control unit receiving the confidence score; And
Including; a lighting device for irradiating the LED light of the specific wavelength according to the signal transmitted from the disease management integrated system, including an LED lighting unit that outputs the LED light of a specific wavelength capable of sterilization; and
The lighting device,
An LED lighting unit provided with an LED for illumination and an LED for sterilization on a lower surface of the substrate, and wherein a lamp shade is provided at an edge of the substrate;
A radiator provided above the substrate of the LED lighting unit to emit heat generated from the substrate to the outside; And
It is provided on the top of the radiator, is connected to the radiator in a state spaced apart from the top of the radiator under the interposition of a spaced connection member, and includes a power supply unit for emitting the lighting LED and the sterilization LED by receiving power through a power line LED sterilization smart barn management system, characterized in that. The method of claim 7,
The LED sterilization smart barn management system, characterized in that the lampshade and the radiator of the LED lighting unit are manufactured by injection molding a synthetic resin material mixed with tourmaline.
The method of claim 7,
LED sterilization smart livestock house management system, characterized in that the upper end of the power supply unit is provided with a moving cradle that is moved along a moving guide frame installed on the upper part of the house. The method of claim 7,
A mounting ring is provided at an upper end of the power supply unit, and the mounting ring is supported by a moving mounting hook that is moved along a moving guide frame installed on the upper part of the livestock house. A sensing step of collecting biological indicator data and environmental indicator data in the livestock house;
A disease that completes a learning model after learning a deep neural network to output a confidence score by calculating the probability of occurrence for each of the representative diseases according to breeding breeds based on the data obtained by normalizing the biological and environmental indicator data. Learning stage;
Disease prediction in which a predicted value for each representative disease is calculated by receiving the deep neural network and inputting the biological and environmental indicator data acquired in the sensing step into the deep neural network, and converting the calculated predicted value into a confidence score for each disease step; And
According to the confidence score and disease type data, it is checked whether the reliability of a specific disease exceeds the threshold value, and when it is exceeded, the LED lighting signal of a specific wavelength value for sterilizing the pathogen of the disease is transmitted by LED light. Lighting control and sterilization steps to sterilize;
LED sterilization smart barn management method comprising a.

Images