Kamalpreet Singh

Bio: Kamalpreet Singh is an academic researcher from Lovely Professional University. The author has contributed to research in topics: Inverter & Low voltage. The author has an hindex of 1, co-authored 4 publications receiving 20 citations.

Hardware Design of Cascaded Multi-Level Inverter using Multi-Sine PWM Technique

Abstract: This paper mainly focusses on the steps for designing cascaded multi-level inverter. The multi-sine pwm technique is implemented which can control the output voltage by varying the amplitudes of two modulating signals. The approach of designing i.e. important features for selecting the controller, bipolar IGBT driver circuit design with an optical isolation, low voltage isolated power supplies for bipolar driver cards and high voltage isolated power supplies for each H-bridge, importance of delay circuit in inverters and design procedure for selecting minimum LC filter size to eliminate harmonics in load as per IEEE-519. The performance analysis and results are verified with MATLAB Simulink and hardware circuit.

Evolution in Solid-State Transformer and Power Electronic Transformer for Distribution and Traction System

Abstract: Power electronic transformer (P.E.T.) and solid-state transformer (SST) are one of the promising technologies in medium and high power conversion systems. In case of controlling power quality for the various load connected, P.E.T. and SST perform greatly in comparison with conventional line frequency transformer (CLFT) With advancements in high power switches and magnetic materials, P.E.T. can reach the efficiency almost equal to CLFT in distribution as well as traction system. P.E.T. eliminates all the limitation that CLFT faces in region of power quality maintenance and power transfer. Over the past two decades, researches and field trial studies are conducted to explore the challenges faced by conventional P.E.T. models and improved them to face all sort of applications in electrical systems. This paper aims to review the essential requirements of P.E.T. and SST modules both in traction and distribution systems. Basic design topologies of both SST and P.E.T. modules are also reviewed. There is also tabulation of all the models recently manufactured by companies for railways and distributions. Finally, this paper discusses the latest go-through in advanced topologies of P.E.T. and their architectural designs.

Induction Motor Internal and External Fault Detection

Abstract: Induction motors are extensively used motor type for various industrial applications for the reason that they are robust, simple in structure, and efficient. On the other hand, induction motors are prone to different faults during their lifetime due to hostile environments. If the fault is not detected in its rudimentary phase, it may cause unexpected shut down of the entire system and colossal loss in industry. It is conspicuous that scope of this field is huge. This work presents detection of internal and external faults of induction motor. S-Transformation, which is superior as compared to CWT and STFT as it does not contain any cross terms, is used for bearing fault detection, and random forest, an algorithm which is easy to implement and requires minimum memory, is used for detection of external faults. The fault can be detected with more accuracy in premature state leads to improve the reliability of the system.

Benchmarking of Document Image Analysis Tasks for Palm Leaf Manuscripts from Southeast Asia

Abstract: This paper presents a comprehensive test of the principal tasks in document image analysis (DIA), starting with binarization, text line segmentation, and isolated character/glyph recognition, and continuing on to word recognition and transliteration for a new and challenging collection of palm leaf manuscripts from Southeast Asia. This research presents and is performed on a complete dataset collection of Southeast Asian palm leaf manuscripts. It contains three different scripts: Khmer script from Cambodia, and Balinese script and Sundanese script from Indonesia. The binarization task is evaluated on many methods up to the latest in some binarization competitions. The seam carving method is evaluated for the text line segmentation task, compared to a recently new text line segmentation method for palm leaf manuscripts. For the isolated character/glyph recognition task, the evaluation is reported from the handcrafted feature extraction method, the neural network with unsupervised learning feature, and the Convolutional Neural Network (CNN) based method. Finally, the Recurrent Neural Network-Long Short-Term Memory (RNN-LSTM) based method is used to analyze the word recognition and transliteration task for the palm leaf manuscripts. The results from all experiments provide the latest findings and a quantitative benchmark for palm leaf manuscripts analysis for researchers in the DIA community.

Study on feature extraction methods for character recognition of Balinese script on palm leaf manuscript images

Abstract: The complexity of Balinese script and the poor quality of palm leaf manuscripts provide a new challenge for testing and evaluation of robustness of feature extraction methods for character recognition. With the aim of finding the combination of feature extraction methods for character recognition of Balinese script, we present, in this paper, our experimental study on feature extraction methods for character recognition on palm leaf manuscripts. We investigated and evaluated the performance of 10 feature extraction methods and we proposed the proper and robust combination of feature extraction methods to increase the recognition rate.

The Handwritten Sundanese Palm Leaf Manuscript Dataset from 15th Century

Abstract: In order to preserve the Sundanese palm leaf manuscripts, some digitization campaigns have been done recently Then, for further access in education and research, the handwritten Sundanese palm leaf manuscript dataset called Lontar Sunda dataset has been created The dataset was constructed from 66 pages of 27 collections of Sundanese palm leaf manuscripts from the 15th century The dataset has been carried out with manuscripts from Garut, West Java, Indonesia This paper presents the Sundanese dataset which is publicly available for scientific use The groundtruth includes binarized images, annotations at word level and annotations at character level The Sundanese dataset provides useful data to test word spotting, character/symbol recognition and binarization methods, and will facilitate the evaluation of developed methods

Effective handwritten digit recognition based on multi-feature extraction and deep analysis

Abstract: Handwritten digit recognition is an important research topic in computer vision and pattern recognition. This paper proposes an effective handwritten digit recognition approach based on specific multi-feature extraction and deep analysis. First, we normalize images of various sizes and stroke thickness in preprocessing to eliminate negative information and keep relevant features. Secondly, considering that handwritten digit image recognition is different from traditional image semantics recognition, we propose specific feature definitions, including structure features, distribution features and projection features. Moreover, we fuse multiple features into the deep neural networks for semantics recognition. Experiments results on benchmark database of MNIST handwritten digit images show that the performance of our algorithm is remarkable and demonstrate its superiority over several existing algorithms.

Letter-based classification of Arabic scripts style in ancient Arabic manuscripts: Preliminary results

Abstract: Classifying ancient Arabic manuscripts based on handwriting styles is one of the important roles in the field of paleography. Recognizing the style of handwriting in Arabic manuscripts helps in identifying the origin and date of ancient documents. In this paper we proposed using segmented letters from Arabic manuscripts to recognize handwriting style. Both Gabor Filters (GF) and Local Binary Pattern (LBP) are used to extract features from letters. The fused features are sent to Support Vector Machine (SVM) classifier. Experimental results have been implemented using manuscripts images from the Qatar National Library (QNL) and other online datasets. Better results are achieved when both GF and LBP descriptors are combined. The recognized Handwritten Arabic styles are Diwani, Kufic, Naskh, Farsi, Ruq'ah and Thuluth.