BUSCA DE PALAVRAS CHAVE EM IMAGENS DE LIVROS IMPRESSOS USANDO VISÃO COMPUTACIONAL

With the increasingly frequent use of books in digital format, people search for the desired subjects in a faster way compared to the search in physical books. This work aimed to develop a computational resource in the form of an application for Android smartphones, which, based on an image captured from a page in a book, performs searches by keywords. The purpose of using the application is to help the reader to find the desired information quickly. We use Computer Vision techniques with the aid of the OpenCV library in the development of algorithms to perform segmentation, correction of the perspective of the book page image, identification and rectification of the wavy lines, recognition and character classification. The results shown were promising with a hit rate of over 88%.

A Deep Learning System for Recognizing and Recovering Contaminated Slider Serial Numbers in Hard Disk Manufacturing Processes

This paper outlines a system for detecting printing errors and misidentifications on hard disk drive sliders, which may contribute to shipping tracking problems and incorrect product delivery to end users. A deep-learning-based technique is proposed for determining the printed identity of a slider serial number from images captured by a digital camera. Our approach starts with image preprocessing methods that deal with differences in lighting and printing positions and then progresses to deep learning character detection based on the You-Only-Look-Once (YOLO) v4 algorithm and finally character classification. For character classification, four convolutional neural networks (CNN) were compared for accuracy and effectiveness: DarkNet-19, EfficientNet-B0, ResNet-50, and DenseNet-201. Experimenting on almost 15,000 photographs yielded accuracy greater than 99% on four CNN networks, proving the feasibility of the proposed technique. The EfficientNet-B0 network outperformed highly qualified human readers with the best recovery rate (98.4%) and fastest inference time (256.91 ms).

Identification of Dendrobium (Orchidaceae) in Liwa Botanical Garden Based on Leaf Morphological Characters

Orchid is one of the most popular ornamental plants in the world. One of the orchid genera that is collected in a large number and known to have high morphological variations in the Liwa Botanical Garden is Dendrobium. However, to date, many Dendrobium collections have not been identified. Given the urgency of identification and the limitations of specimens in the field, especially flower organs, this study is important. This study aims to determine variations in morphological characters, phenetic relationships, and to identify Dendrobium collections based on leaf morphological characters in the Liwa Botanical Garden. Five accessions of Dendrobium were collected, namely CAT140, CAT 144, CAT 271, CAT 274, and IR015. Observation of 11 morphological characters leaves showed that leaf had high variations. The phenetic relationship based on the Gower similarity value and the UPGMA method shows that the Dendrobium in the Liwa Botanical Garden can be classified into 2 main groups formed with a similarity index value of 0.813. Based on Principle Component analysis values, it is known that the characters that have a large influence on grouping are the ratio of leaf length and width, leaf cross section, and leaf arrangement. The phenetic dendrogram topology is supported by the morphological character classification. The results of this study are expected to be basic information in the identification of natural orchids and conservation efforts in the Liwa Botanical Garden.

Optical character recognition system for Baybayin scripts using support vector machine

In 2018, the Philippine Congress signed House Bill 1022 declaring the Baybayin script as the Philippines’ national writing system. In this regard, it is highly probable that the Baybayin and Latin scripts would appear in a single document. In this work, we propose a system that discriminates the characters of both scripts. The proposed system considers the normalization of an individual character to identify if it belongs to Baybayin or Latin script and further classify them as to what unit they represent. This gives us four classification problems, namely: (1) Baybayin and Latin script recognition, (2) Baybayin character classification, (3) Latin character classification, and (4) Baybayin diacritical marks classification. To the best of our knowledge, this is the first study that makes use of Support Vector Machine (SVM) for Baybayin script recognition. This work also provides a new dataset for Baybayin, its diacritics, and Latin characters. Classification problems (1) and (4) use binary SVM while (2) and (3) apply the multiclass SVM classification. On average, our numerical experiments yield satisfactory results: (1) has 98.5% accuracy, 98.5% precision, 98.49% recall, and 98.5% F1 Score; (2) has 96.51% accuracy, 95.62% precision, 95.61% recall, and 95.62% F1 Score; (3) has 95.8% accuracy, 95.85% precision, 95.8% recall, and 95.83% F1 Score; and (4) has 100% accuracy, 100% precision, 100% recall, and 100% F1 Score.