The following specification particularly describes the invention and the manner in which it is to be performed:
TECHNICAL FIELD
[001] The present invention herein relates to the field of agriculture, more particularly detection of diseases in the potato plant through deep learning algorithm and monitor the plant health/growth in real-time with minimal human intervention.
BACKGROUND
[002] In the field of agriculture plant diseases are plays a pivotal role where farmers face tremendous loss due to the failure to detect any diseases occurring in the plants. This leads to damage in the crops as well as the produce and the farmer is unable to make profit during harvest.
[003] There are various common diseases of potatoes which includes ‘potato pest', 'ring spot', 'potato yellowing leaf roll type disease', 'potato wilt' and 'potato anthracnose' . These diseases are more common to occur on the plant due to various reasons such as temperature, soil moisture, pH level etc. These diseases could ruin the plant health and cause damage to the product which would bring greater loss to the farmers.
[004] One needs to pay attention while growing the crops and monitor the plant health in real-time which could assist in prevention of the disease and provide assistance for the growth of the plant with minimal human intervention.
[005] US10349584B2 discloses an invention of a system and method for automatic plant monitoring through machine vision. However the invention fails to disclose that it can only monitor the plant through tests performed and lacks in accuracy.
[006] Potato Plant Leaves Disease Detection and Classification using Machine Learning Methodologies by Aditi Singh and Harjeet Kaur 2021 IOP Conf. Ser.: Mater. Sci. Eng. 1022 012121 discloses a methodology to identify the diseases and classification of diseases that occur for the potato plants. However the invention fails to disclose that it is a framework to only detect the diseases in plant leaves through image processing assisted machine learning technique.
[007] AU2021101682A4, discloses an invention of Automatic plant leaf disease diagnosis with machine learning and deep convolutional neural networks. However, the invention fails to disclose that it is system for diagnosis of plant leaf species diseases only through pre-trained dataset.
[008] CN110033015A, discloses an invention of a kind of plant disease detection method based on residual error network ResNet method. However, the invention fails to disclose that it is a complex and expensive method for the farmers to implement in the farming land for detection of diseases.
[009] However, the present system fails to disclose the drawbacks of the existing system where the plant diseases in not diagnosed based on the real-time plant or failure to prevent the diseases by taking immediate measures based on the potato plant health. 20130211977A1
[0010] The above mentioned prior art states that there is a need for a system or device which provides accuracy in detection of the potato plant health with real-time analysis which could detect any diseases in the potato plant.
[0011] The present invention addresses the above mentioned short comings of the prior art.
SUMMARY
[0012] The summary as given below.
[0013] The present invention is real-time vision assisted real-time health monitoring system designed with deep learning algorithm to detect the diseases in a potato plant .The system monitors various plant health parameters to prevent diseases and promote the growth of the plant through wireless communication .
[0014] In one implementation, pluralities of high-definition cameras are deployed in the farm to capture the real-time images of the plant at regular intervals to monitor without human intervention.
[0015] In one implementation, plurality of sensors in the sensory node such as soil moisture sensor , temperature and humidity sensor , and strain gauge sensor deployed in the system.
[0016] In one implementation, generates real-time alerts via notification upon detection of disease in the plant for immediate assistance to the user.
[0017] In the present implementation, a mobile computing unit , provides access through
[0018] In the present implementation, wireless communication which includes through for effective communication to the cloud server via internet.
[0019] In one implementation, a deep learning algorithm, pre-trained with potato plant diseases classified dataset to detect and identify the diseases occurred in the plant.
[0020] In the present implementation, a motor along with a water pump is deployed near the plant attached with a sprinkler to maintain the soil moisture.
[0021] In one implementation, a Wi-Fi modem deployed to establish wireless communication for transfer of real-time data to the computing unit from the farm.
[0022] In one implementation, provides greater accuracy in detection of diseases in the potato plants through real-time analysis with assistance of a wireless anemometer.
[0023] In the present implementation, monitors the plant health at regular interval without any human intervention and assists in the growth of the plant.
[0024] In one implementation, provides remote access to the user through wireless communication with authenticated login credentials on the webpage.

BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The foregoing detailed description of embodiments is better understood when read in conjunction with the appended drawings. For illustrating the disclosure, example constructions of the disclosure are shown in the present document; however, the disclosure is not limited to the specific methods and apparatus disclosed in the document and the drawings.
[0026] The detailed description is given with reference to the accompanying figures. In the figures, the left-most digit (s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the drawings to refer like features and components.
[0027] Figure 1, illustrates, a block-diagram of the present invention.
[0028] The figures depicts an embodiment of the present disclosure for the purpose of illustration and understanding only
DETAILED DESCRIPTION
[0029] The disclosed embodiments the foregoing detailed description of embodiments is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the disclosure, example constructions of the disclosure are shown in the present document; however, the disclosure is not limited to the specific methods and apparatus disclosed in the document and the drawings.
[0030] In one of the embodiments the detailed description is given with reference to the accompanying figures. In the figures, the left-most digit (s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the drawings to refer like features and components. The disclosed embodiments are mere exemplary of the disclosure, which may be embodied in various forms.
[0031] Some embodiments of this disclosure, illustrating all its features, will now be discussed in detail.
[0032] In one implementation, the vision based artificial intelligence algorithm (120), which performs automated image capturing, image process and analysis of the plants in the farm through a camera (101) deployed at various location of the farming land.
[0033] The disclosure herein is an automated system for diseases detection in potato plant through deep learning is designed to detect plant diseases and monitor the plant health automatically without human intervention.
[0034] In one embodiment, the system deployed with on-board pre-trained deep-learning algorithm which is set in classification of standard based on with various images of the plant diseases in a potato plant along with the normal plant to detect and identify the diseases.
[0035] In the other embodiment, the controller unit (100), main processor unit which is interfaced to the sensor nodes and camera (101) which receives the input and controls the operation of the system by transmitting to the cloud server (104) through wireless communication (103).
[0036] In another embodiment , the plurality of high-definition cameras(101) are deployed in the farms to capture the real-time images of the plant which provides assistance to monitor the potato plant health and detect any diseases which might occur
[0037] In other embodiment, the mobile computing unit (105) allows access through user login via an authenticated username and password to view the details of the potato plant health in real-time through wireless communication.
[0038] In another embodiment, the cloud server (104) receives the sensory data of the plant along with the real-time images capture by the plurality of camera which creates a cloud database for user access.
[0039] In another embodiment ,the motor (11) attached with water pump (12) for water flow through the water sprinkler (130) is enabled automatically through the control unit (100) via the relay (13) upon detection of moisture level in the soil based on the sensory data of the soil moisture sensor (10).
[0040] In other embodiment, the sensor nodes (107) which consists of soil moisture sensor (15) which monitor the moisture content in the soil , temperature and humidity sensor (16) which monitor the appropriate temperature and humidity of the plant , and strain gauge sensor (18) for monitoring the pressure applied is interfaced to the control unit (100) for detection of various parameters of the potato plant.
[0041] In the present embodiment, the wireless anemometer (17), provides accuracy in wind speed, direction, temperature and air pressure for any unforeseen circumstances due to change in weather which sends alert to the user for immediate assistance.
[0042] In other embodiment , the system comes with wireless communication which may include through Wi-Fi modem for transfer of real- time data from the control unit (100) to the cloud server (104) for creation of the cloud based database which is accessible by the user remotely.
[0043] In another embodiment, the mobile computing (105) receives the data from the system through the cloud server (104) via wireless communication (103) such as Wi-Fi or mobile internet.
[0044] In the present embodiment, the system generates an alert to the user upon detection of any abnormal pattern of the sensory data which exceeds the threshold level or upon detection of the any abnormal pattern of the images which are pre-processed through the deep learning algorithm.
[0045] In the other embodiment, the camera (101) upon capturing the raw images of the potato plant is transferred to the pre-processing unit (102) in which the real-time images are processed to remove noise or image distortions and sends the processed image to the deep learning algorithm unit (108) for further analysis of disease detection.
[0046] In the present embodiment, the pre-trained dataset of potato plant is pre-fed into the classifier (109) which consists of various potato plant diseases such as potato pest', 'ring spot', 'potato yellowing leaf roll type disease', 'potato wilt' and 'potato anthracnose' to detect the disease of the potato plant heath.
[0047] In another embodiment, the power supply (115), provides power supply to the system through a solar powered rechargeable batteries (14) for energy conversation.
[0048] Exemplary embodiment
In an exemplary embodiment, is an automated system for diseases detection in potato plant through deep learning is designed to detect plant diseases and monitor the plant health automatically without human intervention. The system monitor the plant health through a sensory node (107) which consists of various sensors such as of soil moisture sensor (15) which monitor the moisture content in the soil, temperature and humidity sensor (16) which monitor the appropriate temperature and humidity of the plant , and strain gauge sensor (18) for monitoring the raindrops on the plants, interfaced to the control unit (100). The system is deployed with the wireless anemometer (17), provides accuracy in wind speed , direction , temperature and air pressure in the farm. The system comes with pre-trained deep learning algorithm which detects the diseases in the plant based on the pre-fed classifier images. A plurality of high definition camera (101) is deployed at various locations of the farm to capture the real-time images of the potato plant. The pre-processing unit (102) receives the raw images capture by the camera (101) and processing by removing unwanted noise or any distortions in the images by providing images with clarity. The classifier unit (109) having a collection of the dataset of the potato plant images which classifies the various plant diseases and assists the deep-learning algorithm unit (108) to detection as well as identification of the potato plant diseases accurately. The sensory node (107) interfaced to the controller unit (100) to monitor the plant health at regular interval. The motor (11) for the water pump connected to the sprinkler (14) is interfaced to the controller unit (100) is enabled through a relay (13) upon detection of soil moisture through the sensory node. The system established communication through wireless communication and transmits the sensory data along with the images to the cloud server (104) through a Wi-Fi modem (112). The mobile computing unit (105) provides access to the user to view the sensory data along with the plant diseases detection through authenticated login credentials via the internet. The system provides power supply (115) through solar powered rechargeable batteries (14) for energy conservation. The system upon detection of any diseases or abnormal data generates an alert (110) to the user through a notification for immediate assistance. The system upon detection of low level of moisture in the soil sends signal to the controller unit (100) which enables the relay (13) to activate the sprinkler which sprays the water on the plant to retain the moisture.

[0049] Referring to figure 1, in which the deep learning unit (108) is pre-fed with algorithm which detects the potato plant diseases based on the real-time images without any human intervention or assistance. The system generates an alert automatically upon detection of the diseases or any abnormal pattern on the potato plant which can assist the farmers to protect the plant from being destroyed.
[0050] Again referring to figure 1, the classifier unit (109) with pre-trained dataset of various images of the potato plants which assists in detection of the diseases in the potato plants.
[0051] Again referring to figure 1, the water sprinkler unit (130), enables automatically upon receiving the signal from the controller unit (100) via the relay (14). The relay (14) acts as switches which enable and disable the sprinkler activation upon detection of low soil moisture based on the sensory node real-time data.
[0052] In further embodiments, may include long range communication, LoRa radio module to provide remote access to the user which could control the operation of the system from long range.
[0053] Some of the embodiments may be further upgraded based upon the study performed further.

We Claim:

1. A real-time monitoring system for detection of diseases in a potato plant through a deep learning algorithm(108) comprises of :
a. a plurality of high-definition camera(101) interfaced to a controller unit (100) deployed in a farm to capture the real-time images of the plant ;
b. a plurality of sensory node (107) deployed on the potato plant;
c. processing of real-time images by a pre-processing unit (102);
d. classification of the potato plant dataset by a classifier unit (109);
e. communication by a wireless communication (103) through internet gateway (106) ;
f. a storage of real-time data by a cloud server (104);
g. a water sprinkler unit (130);
h. access of real-time data in a mobile computing (105) unit by authenticated username and password;
i. generates an alert (110) by notification to the mobile computing unit (105) ; and ,
j. a power supply (115) by a solar powered rechargeable batteries (14).
2. The system as claimed in claim 1, wherein , the sensory node (107) comprises of :
a. a soil moisture sensor (15), evaluate the moisture level in the soil;
b. a temperature and humidity sensor (16), detect the temperature and humidity in the atmosphere;
c. a strain gauge sensor (18), detect the raindrops on the potato plant;
d. a wireless anemometer (17), for evaluation of in wind speed , direction , temperature and air pressure in the farm
3. The system as claimed in claim 1, wherein , the water sprinkler unit (130) comprises of ;
a. a sprinkler head (09), to sprinkle water to the potato plant;
b. a water pump (12);
c. a controller unit (100);
d. a motor (11) 12v Dc motor ; and,
e. a relay (13) as a switch to enable/disable the sprinkler.
4. The system as claimed in claim 1 wherein, the deep learning algorithm (108), detects any diseases in the potato plant based on the real-time images through classifier unit (109) upon receiving from pre-processing unit (102).
5. The system as claimed in claim 1 wherein, monitors the various health parameters of the potato plant based on the sensory node (107) values.
6. The system as claimed in claim 1 wherein, enables the water sprinkler (130) automatically through controller unit (100) upon detection of low soil moisture for potato plant hydration.
7. The system as claimed in claim 1, wherein, the cloud server (104) receives the real-time data of the potato plant through wireless communication for remote access.
8. The system as claimed in claim 1 , provides energy efficiency through the solar powered rechargeable batteries as source of power supply.
9. The system as claimed in claim 1 provides user access to the mobile computing unit via authenticated login credentials for improvised security.
10. The system as claimed in claim 1, upon detection of any diseases as well abnormal pattern in potato plant health immediately generates the alerts (110) by notification for the users immediate attention.