Shanu Sharma

Leaf disease detection and grading using computer vision technology & fuzzy logic

In Agriculture, leaf diseases have grown to be a dilemma as it can cause significant diminution in both quality and quantity of agricultural yields. Thus, automated recognition of diseases on leaves plays a crucial role in agriculture sector. This paper imparts a simple and computationally proficient method used for leaf disease identification and grading using digital image processing and machine vision technology. The proposed system is divided into two phases, in first phase the plant is recognized on the basis of the features of leaf, it includes pre-processing of leaf images, and feature extraction followed by Artificial Neural Network based training and classification for recognition of leaf. In second phase the disease present in the leaf is classified, this process includes K-Means based segmentation of defected area, feature extraction of defected portion and the ANN based classification of disease. Then the disease grading is done on the basis of the amount of disease present in the leaf.

Apple fruit detection and counting using computer vision techniques

In agriculture sector the problem of identification and counting the number of fruits on trees plays an important role in crop estimation work. At present manual counting of fruits and vegetables is carried out at many places. Manual counting has many drawbacks as it is time consuming and requires plenty of labors. The automated fruit counting approach can help crop management system by providing valuable information for forecasting yields or by planning harvesting schedule to attain more productivity. This work presents an automated and efficient fruit counting system using computer vision techniques. The proposed system uses minimum Euclidean distance based segmentation technique for segmenting the fruit region from the input image. Further circle overlaying is done on the fruit region and in the last fruits are counted on the basis of the centroid of the fruit regions. This proposed system is correctly detecting and counting the apples on the test images.

Computer vision & fuzzy logic based offline signature verification and forgery detection

Automated signature verification and forgery detection has many applications in the field of Bank-cheque processing, document authentication, ATM access etc. Handwritten signatures have proved to be important in authenticating a persons identity, who is signing the document. In this paper a Fuzzy Logic and Artificial Neural Network Based Off-line Signature Verification and Forgery Detection System is presented. As there are unique and important variations in the feature elements of each signature, so in order to match a particular signature with the database, the structural parameters of the signatures along with the local variations in the signature characteristics are used. These characteristics have been used to train the artificial neural network. The system uses the features extracted from the signatures such as centroid, height - width ratio, total area, Ist and IInd order derivatives, quadrant areas etc. After the verification of the signature the angle features are used in fuzzy logic based system for forgery detection.

Image processing based degraded camera captured document enhancement for improved OCR accuracy

Over the past decade the document analysis and processing related to camera based document images has gained the interest of research community. Nowadays, cameras are easily available in the smart phones that can be carried in the small space of our pockets while being lightweight, portable and relieving us from the burden of walking down to a scanner for a digital copy of a document. But even though capturing a document image through a phone camera appears simple, the chances of obtaining a perfect picture are scanty. As when the picture is captured in an unconstrained environment, there are chances of degradation to creep in that will hamper the visual quality of the document image which further effect the readability(in terms of OCR accuracy). Low quality documents give poor results. Document images contain various degradations such as blur, uneven illumination, perspective distortion, low resolution, smear etc. Quality enhancement is helpful to recognize a camera captured document more accurately and if not completely removing the degradations, it can be used for suppressing them and making the text more readable. This paper evaluates the performance of various deblurring techniques for noisy and blurred camera captured documents.

An Exploration in Perception-Based Digital Media Processing: A Psychological Perspective

The computer vision field deals with the problem of understanding the scene or features in images of real world with the help of image processing and pattern recognition techniques. The main complication in this task is that the objects present in the images may have different appearances to the camera due to illumination effects, camera position, shadows, types of camera, etc. Nevertheless, with the advancement of technologies, today computer vision has provided reliable methods for various tasks like object classification, action recognition, autonomous driving, scene analysis, highlights extraction in videos and many more. But the problem of automatic qualifying is that how well people perform these actions has been largely unexplored. Human visual system and cognition can outperform the performance of computer vision algorithms. The objective of this paper is to highlight the state of the art of various psychological views of human visual perception in computer vision methods that have been found to operate well and that led up to the above-mentioned capabilities.

Review and comparison of face detection algorithms

With the tremendous increase in video and image database there is a great need of automatic understanding and examination of data by the intelligent systems as manually it is becoming out of reach. Narrowing it down to one specific domain, one of the most specific objects that can be traced in the images are people i.e. faces. Face detection is becoming a challenge by its increasing use in number of applications. It is the first step for face recognition, face analysis and detection of other features of face. In this paper, various face detection algorithms are discussed and analyzed like Viola-Jones, SMQT features & SNOW Classifier, Neural Network-Based Face Detection and Support Vector Machine-Based face detection. All these face detection methods are compared based on the precision and recall value calculated using a DetEval Software which deals with precised values of the bounding boxes around the faces to give accurate results.

Extraction and Enhancement of Moving Objects in a Video

Detection of objects for relocation in a video is a vital as well as initial step for many computer vision-based applications like moving object extraction, video surveillance, pattern classification, etc. The traditional methods used for detection of foreground objects include background subtraction, optical flow and frame differencing techniques. These methods are found to be advantageous only if the extraction of the moving object is precise and clearly visible that it is, the object must be of good quality. This paper emphasizes on the detection as well as the enhancement of the foreground objects. The proposed method uses the amalgam of two traditional techniques background subtraction and motion vector-based optical flow method along with morphological operators to extricate the nonstationary objects from the videos followed by the enhancement of the extracted object to be of better quality in terms of visibility. The proposed algorithm is executed over the videos having frame dimension of 640 × 360 along with the frame rate of 30 frames/second using MATLAB R2013.

An analysis of vision based techniques for quality assessment and enhancement of camera captured document images

Now a days with the advancement of technology, there has been a tremendous rise in the volume of captured and distributed content. Image acquisition can be done with the help of scanners, cameras, smart phones, tablets etc. Document retrieval and recognition systems require high quality document images but most of the time, the images acquired suffer from various degradations like blur, uneven illumination, low resolution etc. To reduce the processing time and get good results, we require methods to evaluate and improve the quality of such images. This paper reviews the quality assessment methods and enhancement techniques for document images as well as represents a survey of the work that has been performed in the field of document image quality assessment and enhancement.

Analysis of computer vision based techniques for motion detection

Motion detection or moving object detection in a video is a process of identification of the change of physical position of the object. Nowadays the field of computer vision based motion detection has attracted many researchers due to its vast applications in the field of intelligent video surveillance, traffic monitoring, event detection, people tracking and behavior analysis etc. Four basic approaches background subtraction, frame difference; temporal difference and optical flow estimation based motion analysis have been discussed and analyzed in terms of their usability and computational time in this paper. Further the various related work under these approaches have been discussed and analyzed. The review presented in this paper is an attempt to provide the thorough understanding of various traditional approaches for motion detection and to present the current scenario of research in this area.

Analysis and Optimization of Feature Extraction Techniques for Content Based Image Retrieval

The requirement of improved image processing methods to index increasing image database that results in an alarming need of content based image retrieval systems, which are search engines for images and also is an indexing technique for large collection of image databases. In this paper, an approach to improve the accuracy of content based image retrieval is proposed that uses the genetic algorithm, a novel and powerful global exploration approach. The classification techniques—Neural Network and Nearest Neighbor have been compared in the absence and presence of Genetic Algorithm. The computational results obtained shows the significant increase in the accuracy by incorporating genetic algorithm for both the classification techniques implemented.