The information of interest to any geospatial application requires being extracted from Topographic maps (TMs). Extracting information from topographic map represents one of the major bottlenecks due to complex distribution of geographic elements and highly interconnected nature of map features. The need for automated topographic map understanding arose because current methods for the information extraction are not adequate and an automated method is necessary in terms of time and economic efficiency.This work reports on an implementation of Topographic map understanding system to extract spatial information and to provide this information in machine-readable data formats to preserve the digital repository. This data can be used for analytical purposes required in generation of. The paper presents Indian topographic map understanding system (ITMUS) that is characterized by the human mentation and learning capabilities. The ITMUS is comprised of image processing routines, Structure feature descriptors, and adaptive Neuro-fuzzy inference system. The Fuzzy inferencing has been implemented using Sugeno model which utilizes the initial crude domain knowledge about the map legends. Further, system has been trained for various sample regions selected from Open Series Map (OSM) Indian topographic maps. Results of implementation are evaluated against reference data of Survey of India and manual recognition. It has been found that the overall recognition rate of the system is 90.91%. Further, the system’s overall accuracy is determined to be 92.77%.

The prediction of the personality of an individual is a critical problem in both areas whether it is considered in the context of organizations or in the case of our daily lives. Prediction of personality depends on many factors and these factors may vary from one individual to another.Personality prediction is identifying the personalities of individuals through their actions in different situations and observing their behaviours in various circumstances. Personality traits show the different characteristics of different people based on their thoughts, feelings, and behaviours. There can be positive as well as negative personality traits. Personality traits are based on the Big Five Model also known as the OCEAN model i.e. Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. In the previous study, many investigations has been done. They have used different techniques and different algorithms to predict the personality of different people. Some have used handwriting to predict personality using the GSC algorithm. Facial expressions have been used in some studies using CNN features. Few studies have focused on the social networking sites for personality prediction by examining an individual’s reaction to different posts, their comments, their posts, etc. One study predicted personality using AU, LF, POS, Emotional features and their combinations. Apart from this, there are few limitations in these single models discussed above. They work efficiently for only a small dataset but on increasing the size of the dataset their accuracy keeps decreasing. Multimodal is effective in this case and to make the task automatic, an intelligent multimodal agent can identify personality traits better based on both verbal and non-verbal features.

Image captioning is one of the most recent challenges that caught the interest of the computer vision community as well as the Natural Language Processing community. Recently, the tedious task of image captioning has attained quite notable progress by using numerous techniques. The primary goal of this paper is to study existing Deep Learning techniques for Image Captioning. We have discussed a convolutional neural network-based Image Caption generation model and the salient steps involved in it. We have also discussed dataset and evaluation metrics widely used in fundamental systems.

Credit card plays a significant standard in the present wealth. It turns into a necessary piece of the family unit, business, and worldwide exercises. Although utilizing credit cards gives might profits when used carefully and dependably, huge credit and monetary effects might be imported by deceitful practices by fraudsters attributable to the ubiquity of electronic asset moves. Financial institutions try to enhance continuously their fraud detection systems, but fraudsters are at the same time hack into the systems with new techniques and tools. Such cheats cause a danger to the protection of humankind, bringing about monetary misfortunes. There is a requirement for planning progressed extortion discovery answers to limit the perils of these fakes. For the detection of deceits, many machine learning algorithms can be utilized. This note paper first discusses the statistics of credit card frauds in the world and primarily in India, then the type of frauds and gives a diagram to analyze the presentation of a few machine learning algorithms by doing a relative report that can be utilized for classifying transactions as misrepresentation or a real one. It also mentions the currently used state of the art techniques to counter these attacks and highlights its limitations along with proposing a solution for it.

Modern medicine has become reliant on medical imaging. The application of computer has proved to be an emerging technique in medical imaging and medical image analysis. Each level of analysis requires an effective algorithm as well as methods in order to generate an accurate and reliable result. Different modalities, such as X-Ray, Magnetic resonance imaging (MRI), Ultrasound, Computed tomography (CT), etc. are used for both diagnoses as well as therapeutic purposes in which it provides as much information about the patient as possible. Medical image processing includes image fusion, matching or warping which is the task of image registration. Medical image analysis includes Image enhancement, segmentation, quantification, registration, which is the most predominant ways to analyze the image. There are various difficulties in medical image processing and subsequent stages like image enhancement and its restoration; segmentation of features; registration and fusion of multimodality images; classification of medical images; image features measurement and analysis and assessment of measurement and development of integrated medical imaging systems for the medical field. In this paper, the techniques used in medical image analysis have been reviewed and discussed extensively. The vital goal of this review is to identify the current state of the art of medical image analysis methods as a reference paradigm in order to accelerate the performance of existing methods.

Numerous strategies are executed for the identification of peculiarities on the framework. Irregularities based strategies are looking at as proficient from that client purpose based methodology is favored for the Usage of oddity recognition. Presently multi day's decent variety of abnormality strategies are accessible In view of this, it is difficult to think about these strategies. To know this, diverse abnormality Identification is checked on and make a nitty-gritty examination in this. This paper contains examination consider of various oddity discovery strategies. Interruption perception has gained a wide consideration and turns into a gainful field for different looks into, and as yet is the subject of all-inclusive intrigue by specialists. The interruption recognition network still stands up to troublesome circumstance even after numerous long periods of research. Decreasing the tremendous number of wrong cautions all through the procedure of recognizing obscure assault designs stays vague issue. In any case, different research results as of late have appeared there are potential answers for this issue. Inconsistency identification is a key issue of interruption recognition in which Irritations of ordinary conduct determine an appearance of planned or unintended impact assaults, flaw, deformities, and others. This paper displays an outline of research bearings for applying composed and disorderly strategies to handle the issue of inconsistency discovery. The references referred to will cover the huge hypothetical issues lead the analyst in fascinating examination bearings.

The work presented in this paper is related to symbols and toponym understanding with application to scanned Indian topographic maps. The proposed algorithm deals with colour layer separation of enhanced topographic map using k-means colour segmentation followed by outline detection and chaining, respectively. Outline detection is performed through linear filtering using canny edge detector. Outline is then encoded in a Freeman way, the x-y offsets have been used to obtain a complex representation of outlines. Final matching of shapes is done by computing Fourier descriptors from the chain-codes; comparison of descriptors having same colour index is embedded in a normalized scalar product of descriptors. As this matching process is not rotation invariant (starting point selection), an interrelation function has been proposed to make the method shifting invariant. The recognition rates of symbols, letters and numbers are 84.68, 91.73 and 92.19%, respectively. The core contribution is dedicated to a shape analysis method based on contouring and Fourier descriptors. To improve recognition rate, obtaining most optimal segmentation solution for complex topographic map will be the future scope of work.

The goal of the study is to devise an intelligent system to understand topographic map automatically. This paper explains the design of a system to automatically interpret information from scanned Indian topographic map legends set. A method based on perception of shape provides a collective understanding of size, form and orientation as that of human psycho-visual approach, is required towards development of a topographic map legends understanding system. The fundamental of the system are map legend analysis algorithms- Edge detection algorithm and line thinning algorithm to extract patterns and shape features from images of scanned topographic map legends and describe it as primitives which is building entity of shape of legend. An approach is based on feature extraction model and back propagation neural network which allows efficient and coherent management of map legends, recognition processes, recognition results. The system incorporates shape feature and uses back propagation neural network for recognition. The experimental results show that developed system performs well in recognition and understanding of map legends.

Recognizing symbol is the first step in using a topographic map. Despite the prerequisite for extraction of information from topographic map, automated understanding of symbols is a challenging task. The objective of this paper is to explain the development of a system for automatic understanding of symbols from the Indian topographic map. The system has been developed making use of shape analysis method in which complex valued chain coding has been used for representation of the exterior boundary of the shape of the symbol. Fourier discrete transform and Auto-correlation function have been used to define shape descriptors. Classification and recognition have been implemented through template matching method and Similarity measures. The system is trained with 150 samples of each of 20 types of symbols from National digital topographic database (NTDB) for OSM of Indian topographic maps. The developed system is tested for 200 samples of each type of symbol from NTDB. It is found that 84.68% of symbols are understood correctly by the developed system. However, there are some inherent limitations in understanding the symbols from an actual map.

The objective of this paper is to explain the design of a system to automatically interpret information from scanned Indian topographic map legends set. The fundamental of the system are image analysis algorithms- Edge detection algorithm and line thinning algorithm to extract pattern and shape features from images of scanned topographic map legends. The recognition is based on feed forward back propagation neural network. The system is implemented in Java and back end provided for an application is XML file. The configurable sliders provided in application allows for efficient and coherent management of map legends, recognition processes, recognition results. The system incorporates shape feature and uses back propagation neural network for recognition. The result gives 93.75% of accuracy.