for spraying pesticide will get active and spray automatically. So pest can be controlled efficiently by the system without the farmer. When the proposed algorithm detect weed, it sends the signal to beaglebone black and farmer can get the alert via sms to control the weed by themselves only. In case of fire in the field, the proposed algorithm detect the region of fire and send the signal to beaglebone black so that the sprinkler for distributing water to the field gets active and take action to resolve it. Thus pest, weed and fire can be detected and controlled efficiently. Proximity sensor is also playing an important role to find any intruder. In the present of any animal in the field like cow or goat, the sensor sends the signal to beaglebone black and automatically it will send it to the corresponding device to uplift the iron railing from all the four sides of the field. Thus, no animal except bird can enter the field. All the device is getting active by utilizing the eco-friendly energy source of solar panel. Farmers get alert of every above mentioned condition via sms. For future records, the processed data from beaglebone black also get uploaded in cloud storage and an interface can be used by the farmer to get all the records of the field anytime from anywhere.
2.5.1 Pest or Weed Detection Process
The continuous image of the field is captured by the camera in every millisecond and gets processed for any unwilling pattern from known pattern. HOG (Histogram Oriented Gradient) image processing strategy is applied here to distinguish between the known and unknown pattern. The known image pattern like crop leaf are treated as training image set which is input to the image processing algorithm. Then by image comparing technique, classifier classifies the image as known or unknown. If any unknown pattern is observed moveable then it will be treated as a weed and unknown pattern is observed not movable then it will be treated as pest.
Figure 2.4 Proposed image processing method to detect pest and weed.
The same image processing method is used in the integrated Agro-IoT system to detect weed and pest and finally store the data in image database. Figure 2.4 illustrates the internal image process method used in the proposed integrated Agro-IoT system.
2.5.2 Fire Detection Process
The researchers have proposed an image processing technique to detect the flame and detect the fire region. Figure 2.5 illustrates the flow chart of the proposed method which have used in the proposed integrated Agro-IoT system to detect the fire captured through the camera.
2.6 Hardware Component Requirement for the Integrated Agro-IoT System
2.6.1 Sensors
The integrated Agro-IoT system uses different sensors which had been discussed in Table 2.1.
2.6.2 Camera
The integrated Agro-IoT system uses a night vision camera which has zooming capacity and will capture the image of the field in every millisecond. No need of having SD card inside the camera as will transfer the images directly to image databases.
Figure 2.5 Proposed image processing method to detect fire region.
2.6.3 Water Pump
The water pump pumps water from the water reservoir and fill the field with water as need.
2.6.4 Relay
The integrated Agro-IoT system uses a relay to open or close the circuit as per the requirement for different operations. It basically acts as a switch.
2.6.5 Water Reservoir
The Water Reservoir stores water from different sources for watering the field when require.
2.6.6 Solar Panel
The integrated Agro-IoT system uses solar panel to use solar energy for running water pump, camera, beaglebone black and GSM module.
2.6.7 GSM Module
The integrated Agro-IoT system uses a GSM module to establish a connection between beaglebone black and the GSM–GPRS enabled mobile system.
2.6.8 Iron Railing
The integrated Agro-IoT system uses iron railing surrounding the total field to prevent the crops from intruders like goat, cow, etc.
2.6.9 Beaglebone Black
The integrated Agro-IoT system use Beaglebone Black, a small stand-alone linux computer. Here used as an embedded system. Figure 2.6 illustrates the model of beaglebone black.
Figure 2.6 Beaglebone black.
2.7 Comparative Study Between Raspberry Pi vs Beaglebone Black
2.7.1 Raw Comparison
To get Quick Overview of each.
| Specification | BeagleBone Black | Raspberry Pi | Result |
| Processor | 1 GHz TI Sitara AM3359 ARM Cortex A8 | 700 MHz ARM1176JZFS | BeagleBone Black Winner |
| RAM | 512 MB DDR3L @400MHz | 512 MB SDRAM @ 400MHz | BeagleBone Black Winner |
| Storage |
2 GB on-board eMMC, MicroSD
|
