29 research outputs found
Improved Pattern Matching Algorithm
No full textPattern matching problem aims to search the most similar pattern or object by matching to an instance of that pattern in a scene image. In order to address the issue of finding an object in the target image efficiently, the most distinctive features are computed from the query pattern and need to be searched in the scene image. The scene image is logically divided into a number of candidate windows which are then to be matched with the query pattern. Due to repeated matching of the query pattern with local candidate windows, the pattern matching process requires a large amount of space in memory as well as it needs to be executed fast. Thus, pattern matching algorithms need to be memory efficient and as fast as possible. This paper makes an attempt to deal with these issues by presenting two effective pattern matching algorithms, namely, strip subtraction and strip division. The efficacy of the proposed pattern matching algorithms is tested on two databases, viz. a local database and MIT-CSAIL database containing random objects. The experimental results are proved to be computationally efficient ones while the proposed algorithms are compared with some existing algorithms possessing a uniform experimental setup
HPV guided object tracking: Theoretical advances on fast pattern matching technique
No full textSummaryPattern matching is a fundamental machine vision problem that deals with searching an object in a comparatively large scene. It can use to solve many vision problems ranging from typical human detection to searching defective parts in industrial automation. This paper reports a fast pattern matching technique which makes use of cumulative subtraction and cumulative division operations based on Image Integral model. The idea is to use both the cumulative subtraction and division operations to evaluate the image values on a very small rectangular region of the image scene as well as on the input pattern to be searched for. Image values are transformed to Haar Projection Values (HPVs) using Haar transform in order to achieve pattern matching on sliding window of the image scene. Computation of HPV needs seven arithmetic operations, including two addition and five subtraction operations, which are found to be same as that of Image Integral technique. Besides, the proposed pattern matching technique is identified as computationally effective in terms of both time and memory
Searching a pattern in token scene image via multi-variant symmetric pattern matching technique
No full text
Fingerprint matching using graph structure based symmetric ternary pattern
No full textCreative Commons licenseFingerprint matching, one of the sophisticated biometric authentication techniques, is popular for its easy implementation,
persistent nature of the fingerprint and non-similarity nature of two fingerprints. Uniqueness of fingerprint is characterized by
distinctive features present in fingerprint image. This paper presents a novel relational descriptor based fingerprint matching process using pattern matching concept called Multi-Variant Symmetric Ternary Pattern (MVSTP). Orientation and illumination invariant local descriptor MVSTP extract distinct features from fingerprint image by referring non-overlapping neighbor pixels in symmetric way with respect to source pixel positioned at the center of 5×5 pixel area. After feature extraction from query fingerprint image and stored fingerprint images in the database, features are compared to find similarity match. MVSTP aims to increase fingerprint matching accuracy in contrast with other processes by addressing challenges related to fingerprint pattern’s appearance variation with slight orientation and the variations present in image properties. The computational proficiency of the proposed fingerprint matching process is tested on FVC 2004 database and local database of fingerprint images with higher note of matching accuracy, manifesting its intensity in the process
Intel 8085 Microprocessor Simulation Tool OneX Simulator
No full textIn spite of the advancement in computer architecture and availability of microprocessors (ex. Intel Core i7 etc.) with speed thousand times greater, microprocessor 8085 is still widely used in academia for education and research purposes. To make the microprocessor 8085 more accessible and portable, many simulators for the same have been introduced over the past years with various level of user friendliness. This paper proposes a simple yet powerful enhanced simulator, namely OneX which is capable of real time code parsing, error handling and label parsing for different addresses. Proposed software also addresses few flaws present in the earlier version of 8085 microprocessor software. The latest copy of the software is available at (https://github.com/Pronoy999/Project-OneX
Learning Cross-Modal Deep Embeddings for Multi-Object Image Retrieval using Text and Sketch
No full textThis is the author accepted manuscript. The final version is available from IEEE via the DOI in this recordIn this work we introduce a cross modal image retrieval system that allows both text and sketch as input modalities for the query. A cross-modal deep network architecture is formulated to jointly model the sketch and text input modalities as well as the the image output modality, learning a common embedding between text and images and between sketches and images. In addition, an attention model is used to selectively focus the attention on the different objects of the image, allowing for retrieval with multiple objects in the query. Experiments show that the proposed method performs the best in both single and multiple object image retrieval in standard datasets.European Union Horizon 2020CERCA Programme/Generalitat de Cataluny
Empowering Medical Education: Unveiling the Impact of Reflective Writing and Tailored Assessment on Deep Learning
No full textIntroduction: Reflective thinking offers learners insight and encourages deeper understanding by leveraging past experiences. This study explores the impact of reflective writing, a selfassessmenttool, on undergraduate medical students. The focus is on training students using author-specific reflection rubrics based on Moon’s model.Methods: A mixed-methods study involving 32 volunteered students undertaking an interactive 3-hour session on reflective thinking and writing (RT&W). 19 students submitted reflections, which were self-graded by students and two faculties independently. The perceptions of students were gathered through questionnaires and focus group discussions. The analysis was done using the mean, inter-class correlational coefficient, and thematic analysis.Results: Inter-rater reliability and inter-class correlation coefficient for reflective writing rubric scores was 63.2%, i.e. below the acceptable threshold. Cronbach’s Alpha for the learner perception questionnaire was 0.90. The outcome of the student’s perception questionnaire recognized the value of reflective writing in terms of professional skills enhancement (4.83±0.39) and improvement after feedback (4.17±0.72). However, satisfaction with overall training was comparatively lower (2.5±0.52). Focus group discussions revealed six themes.Conclusion: Reflective writing enhances the learning outcomes, deepens understanding, and refines judgment. The author-specific reflection rubric, though reliable, warrants empirical validation with a larger and more diverse participant pool. Undergraduate programs should prioritize mastery of reflection and metacognitive learning approaches to optimize educational outcomes
Closed‐loop one‐way‐travel‐time navigation using low‐grade odometry for autonomous underwater vehicles
No full text© The Author(s), 2017. This article is distributed under the terms of the Creative Commons Attribution License. The definitive version was published in Journal of FIeld Robotics 35 (2018): 421-434, doi:10.1002/rob.21746.This paper extends the progress of single beacon one‐way‐travel‐time (OWTT) range measurements for constraining XY position for autonomous underwater vehicles (AUV). Traditional navigation algorithms have used OWTT measurements to constrain an inertial navigation system aided by a Doppler Velocity Log (DVL). These methodologies limit AUV applications to where DVL bottom‐lock is available as well as the necessity for expensive strap‐down sensors, such as the DVL. Thus, deep water, mid‐water column research has mostly been left untouched, and vehicles that need expensive strap‐down sensors restrict the possibility of using multiple AUVs to explore a certain area. This work presents a solution for accurate navigation and localization using a vehicle's odometry determined by its dynamic model velocity and constrained by OWTT range measurements from a topside source beacon as well as other AUVs operating in proximity. We present a comparison of two navigation algorithms: an Extended Kalman Filter (EKF) and a Particle Filter(PF). Both of these algorithms also incorporate a water velocity bias estimator that further enhances the navigation accuracy and localization. Closed‐loop online field results on local waters as well as a real‐time implementation of two days field trials operating in Monterey Bay, California during the Keck Institute for Space Studies oceanographic research project prove the accuracy of this methodology with a root mean square error on the order of tens of meters compared to GPS position over a distance traveled of multiple kilometers.This work was supported in part through funding from the Weston
Howland Jr. Postdoctoral Scholar Award (BCC), the U.S. Navy's Civilian
Institution program via the MIT/WHOI Joint Program (JHK),W. M.
Keck Institute for Space Studies, and theWoods Hole Oceanographic
Institution
