This work, at first, discussed prevalent techniques and methodologies used for the detection, classification and quantification of diseases, respectively. We explained the major phases of plant health monitoring but particularly focused on the detection and classification of spot and blight diseases affecting various plants. Since the visual distinction between both the diseases is based on lesions, it is quite hard to distinguish between them and thus we attempted to provide a prescription for the same. This work has covered diseases of important crops like olive, tomato, corn, grape, and potato, respectively. It is worth mentioning that our work involved the generation of leaf datasets from local olive plantations as well from the leaf image public dataset of PlantVillage.

We have developed the detection and classification techniques for the infected leaves belonging to above- mentioned crops. The major focus of the work was on individual lesions on the leaves as it was less explored in the recent literature and was clearly a gap to work upon. For isolation of lesions, various region of interest segmentation techniques such as k-means segmentation, delta-E color segmentation

and color space thresholding were identified and utilized. Since these lesions are varying in shape, color and texture it's hard to come up with one universal technique. But we identified the most optimal technique for each crop type in this process. For the identification of the features, various texture features were derived from the grey-level co-occurrence matrix. Since the focus was on lesions on

leaves due to various diseases it was interesting to find correlations between some features and the lesions. For classification, a novel classification technique has been successfully developed based upon the statistical analysis of features and their roles in the classification of diseases in various crops.

The software system is developed to automate a BioChemical process of DNA sequencing. The whole system has many modules one of which involves controlling of the 10 port valco valve. To control this device an api is needed which further is integrated into a GUI application made in Qt. This software has a GUI which enables the operator to control which valve to select and hence which liquid to inject on the imagechip. This software controls the labsmith electronic valves and the valco electronic valves. This software makes use of boost libraries, lua scripting, and the development language used is c++ , GIT is used for version control.

After the image is taken from camera on the board it is analyzed using a software which is used to detect the number of beads present in an image and hence the number of DNA molecules attached to the beads, by counting the number of beads one can determine which base is detected and hence can carry on with sequencing process. For detection of beads on the image chip openCV libraries are used. Bead detection algorithms are used to detect the beads in large quantities and also on a stream of images. A GUI based program is made to detect the beads including the options of adding the particular files and the directories. There are two stages of bead detection, in first stage only one image is used to detect the beads, in the second stage a stream of images are used and the beads are verified in all those. The stream of images are time stamped so that a sequence of images are used properly and verification is done efficiently.

The project deals with the design and development of a Farm Robot which is based on ARM9 processor and developed on mini2440 board. The robot is used for smart inspection of the agriculture fields mainly by sensing the changes in the environmental parameters such as the humidity and temperature .Robot is having a mounted camera on top for detection of deformities in certain crops which is detected by a devised image processing algorithm .The project also constitutes a robot controlling and monitoring GUI which is developed in Qt and for the transmission and control, zigbee modules are used for them. Device drivers are also developed and kernel for the board is also modified according to the need of the project and the upgraded hardware

Linux with cross-platform application development framework Qt is becoming popular to develop applications in the domains of agriculture, medicine, and robotics. This project deals with the low cost implementation of multi channel Reverse Osmosis (RO) System for raw saline water. The developed system on line monitors the RO process parameters such as pressure at different valves, Total Dissolved Solids (TDS) and pH from raw and product water by using Qt tool under Linux along with Data Acquisition (DAS) PISO-813 card. The control algorithms generate an ON/OFF command to the actuators to a solenoid valve unit through PISO-P32A32 DAS card according to the set points based on the measured parameters. There are also provisions to display measured parameters as a trend graphs and history graphs from the stored data in files. The system communicates to host computer or other embedded system through RS232 or Wireless IEEE1451.3 network. This embedded system is developed on an open platform which not only reduces the cost of the software but also reduces the development time.