Research Cell: An International Journal of Engineering Sciences

Amandeep Singh ¹, Maninder Lal Singh ¹

Affiliations
1 Department of Electronics Technology, Guru Nanak Dev University, Amritsar, IN

Abstract

Deciding the criterion for the performance evaluation of a classifier plays a vital role in the selection procedure of a classifier for a certain problem. These criteria empower the researchers to do the selection of a classifiers for effective classifications of unseen data from a range of classifying algorithms. A great number of different measures are currently available for the classification problems based on binary, flat or undistributed data such as in case of images. However in case of hierarchical classifications, where the number of classes to be identified are more than two, the evaluation of a classifier becomes more and more intricate as the classes to be differentiated, are hierarchically attached. The topic of focus of this paper is to provide a knowledge flow which a researcher can use while dealing with such real time based problems where the accuracy and efficiency of a classifier are the major concerns. The problem of interest while discussing the different aspects of various evaluation measures, was the color prediction of paddy crop plant leaf for its health characterization.

Keywords

Machine Learning, Confusion Matrix, ROC, AUC, Cost Curve, Accuracy.

Full Text

   

Harshwardhan Kakkar ¹, Jaspreet Kaur ¹, Amandeep Singh ¹

Affiliations
1 ECE Department, GNDU RC, Fattu Dhinga, IN

Abstract

Globally, wheat is the leading source of vegetable protein in human food, having a higher protein content than other major cereals, maize (corn) or rice. In terms of total production tonnages used for food, India is currently second to wheat as the main human food crop and ahead of maize. Wheat grains detection is the major task of differentiating wheat from other grains (like barley) as well as from impurities. We notice that grain size and other properties of wheat grains which are used to detection of wheat from the mixture. Specifying the quality of wheat manually requires an expert judgment and is time consuming. Sometimes the variety of wheat looks so similar to the other cereals or impurities so that differentiating them becomes a very tedious task when carried out manually. To overcome this problem, Image processing can be used to differentiate wheat according to its quality. This inspection approach based on image analysis and processing has found a variety of different applications in the agro industry.

Keywords

Thresholding, Binarization, Major Axis, Minor Axis, Depth of Field.

Full Text

   

Amandeep Singh ¹, Jagtar Singh Sivia ²

Affiliations
1 Yadavindra College of Engineering, Talwandi Sabo, Punjab, IN
2 Yadavindra College of Engineering, Punjabi University, IN

Abstract

In this paper a compact Rectangular Micro-strip Patch Antenna (RMPA) with single feed is presented. Resonant frequency of proposed antenna is shifted toward lower side by introducing rectangular slot at the upper edge of patch. FR4 substrate having with thickness 1.6 mm is used as substrate material for the design of proposed antenna Performance of antenna in terms of Gain bandwidth is improved by adding a small piece of rectangular patch within the area of the rectangular slot. Antenna size has been reduced 48.9% when compared to conventional RMPA. The return loss of antenna of the antenna at resonant frequencies 3.3 GHz, 4.7 GHz and 6.7 GHz are -15.68dB,-17.71 dB and -33.82 respectively. Conventional square micro strip patch antenna gives return loss -10.91 dB at 7.4 GHz frequency. Ansoft HFSS V13.0.0 (high frequency structure simulator) software is used to investigate the characteristics of this antenna.

Keywords

Microstrip, Gain, Returns Loss, High Frequency Structure Simulator, VSWR.

Full Text

   

Amandeep Singh ¹, Pankush Singla ¹, Navdeep Kaur Khiva ¹

Affiliations
1 C.S.E. Deptt., GZS PTU Campus, Bathinda, IN

Abstract

A honeypot is used in the area of computer and Internet security. It is an information system resource which is intended to be attacked and compromised to gain more information about the attacker and the used tools. It can also be deployed to attract and divert an attacker from their real targets. Compared to an intrusion detection system; honeypots have the great advantage that they do not generate false alerts because no productive components are running on the system. This fact allows the system to log every byte and to correlate this data with other sources to draw a picture of an attack and the attacker. Traditionally honeypots are connected with end clients to detect the uneven behavior of traffic. Activities such as port scanning can be effectively detected by the weak interaction honeypot but many applications such as packet scanning, pattern scanning cannot be detected by weak honeypots. In our research we will propose a strong honeypot mechanism along with intrusion detection system to achieve maximum security in the wireless network. To achieve the objective of our research we placed the honeypot just after the Firewall and intrusion system have strongly coupled synchronize with honeypot. Monitoring will be done at packet level and pattern level of the traffic. Simulation will filter and monitor traffic for highlight the intrusion in the network.

Keywords

Honeypot.

Full Text

   

Navdeep Kaur Khiva ¹, Amandeep Singh ¹, Pankush Singla ¹

Affiliations
1 Deptt. of Computer Science and Engineering, GZS PTU Campus, Bathinda, Punjab, IN

Abstract

Cloud computing is a technique in which data has to be transmitted across the internet from one point to another. Since the data is free of the transmission, there is a high possibility that the data sent across may be lost or be attacked by intruders, so the privacy among the data is to be maintained for secure transmission.

Privacy in cloud computing is the ability of a user or a business to control what information they reveal about themselves over the cloud or to a cloud service provider, and the ability to control who can access that information. Numerous existing privacy laws impose the standards for the collection, maintenance, use, and disclosure of personal information that must be satisfied by cloud providers. Cloud Service Providers can store information at multiple locations or outsource it, then it is very difficult to determine, how secure it is and who has access to it.

Cloud computing is a revolutionary computing paradigm which enables flexible, on demand and low-cost usage of computing resources. Those advantages, ironically, are the causes of privacy problems, which emerge because the data owned by different users are stored in some cloud servers instead of under their own control. The privacy problem of cloud computing is yet to be solved effectively.

In this paper, I reviewed many research papers to check how many privacy preserving architectures exists and how they are preserving the data.

Full Text

   

Amandeep Singh ¹, Sunny Behal ¹

Affiliations
1 Department of Computer Science and Engineering, Saheed Bhagat Singh State Technical Campus, Ferozpur, Punjab, IN

Abstract

Wireless sensor networks is very important field in today's technology and one may concern about the life time of sensors as they have no facility to change the battery of those sensors inside the field. Wireless Sensor Networks are prone to node failure due to power loss. In order to provide reliable service through the network, the network should be self-adjusting and must have adaptable properties as required from time to time. Here in this research we have proposed a new algorithm which is capable of not only to do optimize routing even with that it has the benefit to overcome through pits creating problem around the sink. We have used the Energy Efficient Shortest Path Routing algorithm for routing and multi-hop network to communicate every node with sink and have used Ant colony optimization to determine new position for a sink, so that network will communicate without any problem which generally occurred due to dead nodes around the sink. The performance of our proposed algorithm has been tested on static and mobile sink scenarios with varying speed, and compared with other state-of-the-art routing algorithms in WSN. In this research, we have investigated the impact of sink mobility on network lifetime. In a typical WSN, all the data generated in the network are routed to a static sink. Nodes near the sink tend to deplete faster in their energy which might cause holes in the network thus limiting the network lifetime. With the introduction of mobile sink, the nodes around the sink always changes, thus balancing the energy consumption in the network and improving the network lifetime. We have simulated four routing algorithms in two scenarios; static sink and mobile sink. It has been observed from the results, that the network lifetime has been improved by the proposed algorithm in comparison to other algorithms.

Keywords

WSN, Sink, Network, Routing, Hops etc.

Full Text

   

Amandeep Singh ¹, Hiteshwari Sabrol ², Verma T ³

Affiliations
1 Research Scholar, DAV University, Jalandhar, IN
2 Assistant Professor, DAV University, Jalandhar, IN
3 Assistant Professor, CMR Engineering College, Hyderabad, IN

Abstract

Organizations, which include government agencies, the military, and humanitarian groups, are responsible for providing the most vulnerable individuals with aid and protection during emergencies and disasters. Their rapid decisions are made possible through the information they gather. Their information needs vary depending on their specific roles and responsibilities. In times of crisis, they need factual and timely information, especially when there is a lack of reliable sources such as radio or TV. Due to the increasing number of people using social media platforms and mobile technologies, the general public has gained access to more effective and practical ways to share information. The term multimodal refers to the combination of various computational methods used to analyze the data collected from social media platforms. Some studies show how social media analytics can be used to summarize and curate information related to disasters. This paper discusses the research being conducted in the field of crisis informatics, which is an interdisciplinary discipline that combines the expertise of social science and computing to extract information related to disasters. Due to the availability of social media data, this field is heavily focused on developing effective strategies to use it.

Keywords

Multimodal System, Natural Disaster Management, Social Media, Machine Learning.

Full Text