Accurate extraction of the Region of Interest is critical for successful ocular region-based biometrics. In this direction, we propose a new context-based segmentation approach, entitled Ocular Region Context Network (ORCNet), introducing a specific loss function, i.e., the Punish Context Loss (PC-Loss). The PC-Loss punishes the segmentation losses of a network by using a percentage difference value between the ground truth and the segmented masks. We obtain the percentage difference by taking into account Biederman’s semantic relationship concepts, in which we use three contexts (semantic, spatial, and scale) to evaluate the relationships of the objects in an image. Our proposal achieved promising results in the evaluated scenarios—iris, sclera, and ALL (iris + sclera) segmentations—, outperforming the literature baseline techniques. The ORCNet with ResNet-152 outperforms the best baseline (EncNet with ResNet-152) on average by 2. $$27\%$$ , 28. $$26\%$$ and 6. $$43\%$$ in terms of F-Score, Error Rate and Intersection Over Union, respectively. We also provide (for research purposes) 3191 manually labeled masks for the MICHE-I database, as another contribution of our work.

Electrophoresis is essential in molecular biology, providing critical data for genetic research. However, manual interpretation of DNA band patterns in electrophoresis images, particularly for dominant molecular markers, remains challenging and prone to errors. This study applied YOLO (You Only Look Once), an advanced object detection model, to automate electrophoresis gel analysis. Experiments were conducted using a dataset originally composed of 246 manually labeled electrophoresis images, later expanded to 1,200 images through augmentation. The dataset will also be made available as a contribution of this work. Several YOLO models (v5 to v12) were trained over 2,000 epochs. Processing times ranged from 18 to 24 hours, and average precision (mAP) scores across versions were 92.2%, 94.6%, 94.3%, 94.8%, 94.6%, 93.8%, 95.0 and 90,9%, respectively, with YOLO v11 achieving the highest mAP at 95.0%. Additionally, we developed the 'Gel Analysis APP', a cross-platform tool compatible with Windows, Linux, and MacOS, offering an intuitive interface. This software automates electrophoresis gel reading, generating similarity matrices based on molecular marker presence or absence, simplifying a commonly manual task in genetics labs.

Plant diseases are one of the factors that compro-mise food production goals. Tomatoes are one of the world's most consumed vegetables and are widely affected by various diseases. Tomato cultivation in greenhouses enables continuous production. In this context, this research focuses on identifying diseases in greenhouse tomato cultivation scenarios. For this study, new datasets were created with two image sizes: the Tomato Leaf Image Dataset (TLID) with image sizes of 256 x 256 pixels and 15,256 images, and the Patch Tomato Leaf Image Dataset (PTLID) with patch sizes of 32 x 32 pixels and 227,218 images. Both datasets comprise seven classes, including four types of diseases, two combinations of diseases on the same leaf, and the healthy leaf. Machine Learning techniques have been widely used to identify plant diseases. This work presents two machine learning methods tested with both datasets. In the proposed models, three convolutional neural networks were combined: a customized CNN, VGG19, and Resnet50, along with two voting classification methods using Hard and Soft de-cisions. The evaluation conducted on the datasets demonstrated that using patches significantly improves results, achieving an accuracy of 90.48%. This technique enables the identification of the disease stage.

The plant species classification using leaf images is a challenge due to the lack of annotation, imbalanced classes and similarities in the data representation. For such problems, Siamese Neural Networks (SNN’s) have been used to overcome these bottlenecks in several contexts. In light of this, this work evaluates different architectures trained in Siamese manner for classifying plant species from the leaf image. Besides, we combined features from the intermediate convolutional layers to improve representations. Experiments on the well-known Flavia and MalayaKew databases have shown that the fusion of intermediate features results in a relevant gain in performance.

In this study we classified the songs of the Latin Music Mood Database (LMMD) according to their emotion using two approaches: single-step classification, which consists of classifying the songs by emotion, valence, arousal and quadrant; and multistep classification, which consists of using the predictions of the best valence and arousal classifiers to classify quadrants and the best valence, arousal and quadrant predictions as features to classify emotions. Our hypothesis is that breaking the emotion classification in smaller problems would reduce complexity and improve results. Our best single-step emotion and valence classifiers used multimodal sets of features extracted from lyrics and audio. Our best arousal classifier used features extracted from lyrics and SMOTE to mitigate the dataset imbalance. The proposed multistep emotion classifier, which uses the predictions of a multistep quadrant classifier, improved the single-step classifier performance, reaching 0.605 of mean f-measure. These results show that using valence, arousal, and consequently, quadrant information can improve the prediction of specific emotions.

Bonsai consists of an ancient art which is aimed at mimicking a tree in miniature. Despite being original and popular on the Asian continent, Bonsai has been widespread in several parts of the world. There are many techniques for styling the plants, classifying them in different patterns widely known by people who appreciate this art. In this work, we introduce a new database specially created for the development of research on Bonsai styles classification. The database is composed of 700 samples, equally distributed among the seven following classes: Formal Upright, Informal Upright, Slanting, Cascade, Semi Cascade, Literati and Wind Swept. The classes selected to compose the database were chosen considering the five basic styles and two more styles that have distinct characteristics from the others. The database was created by the authors themselves, using images available on the Pinterest platform, and they were subjected to a pre-processing criteria to remove similar photos and resize them. The baseline results presented here were obtained using deep models (CNN architectures) successfully used to address image classification tasks in different application domains: VGG, Xception, DenseNet and InceptionV3. These models were trained on ImageNet and we used transfer learning aiming to adapt it to the current proposal. In order to avoid overfitting, data augmentation was performed during training, along with the dropout method. Experimental results showed that VGG19 model obtained the highest accuracy rate, reaching 89%. In addition, we used DeconvNet and Deep Taylor methods aiming to find a proper explanation for the obtained results. It was noted that the VGG19 model better captured the most important aspects for the classification task investigated here, with a better performance to discriminate and generalize patterns in the task of classifying Bonsai styles.

Video as well as music are potent means to convey emotions. However, despite their importance in several applications, few works deal with the issue of emotion classification in videos. The main reason is possibly the lack of available databases. In this work we extend the CAL500 database by including music videos, since the CAL500 was originally proposed as an audio-only database. The main rationale here is that the music videos must be official as they were developed to convey the same emotion as the song. After adapting the database, we have extracted audio and video features to perform our computational experiments. Our main result is that there is a complementarity between the audio and video features as the best result was achieved using their combination.

In this work we present a comprehensive set of experiments aiming to perform music genre classification using learned and handcrafted features plus the fusion of them. Handcrafted features were obtained from the audio signal itself, lyrics, chords and spectrogram images extracted from the audio. The rationale behind this investigation is based on the assumption that one can find some complementarity between classifiers created from these different resources. The experimental protocol was conducted on the Brazilian Music Dataset using the artist filter restriction and they confirm the power of non-handcrafted features to perform audio classification tasks. The experimental results have shown a significant complementarity among the handcrafted features for which the evaluated fusion strategies allowed an improvement in the classification accuracy up to 4 percent points. On the other hand, the fusion of learned and handcrafted features provided similar accuracy than the best individual CNN (0.7815).

This work introduces a Brazilian Speech Database (BrSD), a novel dataset freely available created to support the development of speech-based recognition tasks. As far as we know, this is the first Portuguese language based database with these characteristics created and made available to the research community. We also describe experiments accomplished on BrSD exploring its different possibilities of classification tasks, i.e., age group and gender classification. We use four well-known acoustic features extracted directly from the audio signal and one texture-based feature extracted from a visual representation of the audio signal, the spectrogram. We considered three different classification scenarios: each feature individually, early fusion of the features, and late fusion of the features. Experiments were conducted using Support Vector Machine (SVM) and Multi-layer Perceptron (MLP) classifiers. The obtained results showed that SVM classifier achieved the best recognition rates both in early and late fusion scenarios. The best recognition rates achieved were 91.25%, 88.75%, and 80.25% for gender, age group, and age-gender classification tasks, respectively.

The task of learning from imbalanced datasets has been widely investigated in the binary, multi-class and multilabel scenarios. Although this problem also affects hierarchical datasets, to the best of our knowledge, there are no works in the literature that deal with imbalanceness in hierarchical contexts. In this paper we propose metrics to measure "how imbalanced" is a Hierarchical Multi-Label Dataset, in addition to an approach to deal with this imbalanceness using Multi-Label resampling techniques. The proposed technique is based on the conversion of the dataset labels to a strictly multi-label format, applying wellknown multi-label resampling techniques and then converting the dataset back to its hierarchical taxonomy. The experimental evaluation over a highly imbalanced Music Genre Recognition dataset achieved promising results, with an increase of 0.2337 in the Avg-AUROC metric in relation to the original dataset.

In this work we address the automatic music genre classification as a pattern recognition task. The content of the music pieces were handled in the visual domain, using spectrograms created from the audio signal. This kind of image has been successfully used in this task since 2011 by extracting handcrafted features based on texture, since it is the main visual attribute found in spectrograms. In this work, the patterns were described by representation learning obtained with the use of convolutional neural network (CNN). CNN is a deep learning architecture and it has been widely used in the pattern recognition literature. Overfitting is a recurrent problem when a classification task is addressed by using CNN, it may occur due to the lack of training samples and/or due to the high dimensionality of the space. To increase the generalization capability we propose to explore data augmentation techniques. In this work, we have carefully selected strategies of data augmentation that are suitable for this kind of application, which are: adding noise, pitch shifting, loudness variation and time stretching. Experiments were conducted on the Latin Music Database (LMD), and the best obtained accuracy overcame the state of the art considering approaches based only in CNN.

This paper proposes a semiautomatic method to segment similar images according to a reference segmentation by combination of hierarchical segmentation and texture analysis. The method identifies delimited regions from a reference image in a target image hierarchically segmented with a minimum area criterion. Comparison of regions is performed by extracting texture features from the delimited regions by LBP texture descriptor and the Kullback-leibler similarity measure. The method segments images whose context is defined by reference image segmentation, independent of semantic of the images, and showed promising results in relation to its purpose.

Image identification of animals is mostly centred on identifying them based on their appearance, but there are other ways images can be used to identify animals, including by representing the sounds they make with images. In this study, the authors present a novel and effective approach for automated identification of birds and whales using some of the best texture descriptors in the computer vision literature. The visual features of sounds are built starting from the audio file and are taken from images constructed from different spectrograms and from harmonic and percussion images. These images are divided into sub‐windows from which sets of texture descriptors are extracted. The experiments reported in this study using a dataset of Bird vocalisations targeted for species recognition and a dataset of right whale calls targeted for whale detection (as well as three well‐known benchmarks for music genre classification) demonstrate that the fusion of different texture features enhances performance. The experiments also demonstrate that the fusion of different texture features with audio features is not only comparable with existing audio signal approaches but also statistically improves some of the stand‐alone audio features. The code for the experiments will be publicly available at https://www.dropbox.com/s/bguw035yrqz0pwp/ElencoCode.docx?dl=0 .

This work aims to present a system for automatic music mood classification based on acoustic and visual features extracted from the music. The visual features are obtained from spectrograms and the acoustic features are extracted directly from the audio signal. The texture operators used are Local Phase Quantization (LPQ), Local Binary Pattern (LBP) and Robust Local Binary Pattern (RLBP). The acoustic features are described using Rhythm Patterns (RP), Rhythm Histogram (RH) and Statistical Spectrum Descriptor (SSD). The experiments performed were made on a subset of the Latin Music Mood Database, considering three mood classes: positive, negative, and neutral. In the classification step, SVM classifier was used and the final results were taken by using 5-fold cross validation. From several classifiers created performing a zoning strategy along the images, the best individual classifier is that created from the image zone which corresponds to the frequencies from 1,700 Hz to 3,400 Hz, and using the RLBP visual descriptor. In this case, the F-measure obtained is about 59.55%.

In this paper a novel approach for automatic bird species classification is described. The proposed strategy is based on features taken from the textural content of spectrogram images of bird vocalizations. We show how several texture descriptors can be used for representing the spectrograms. The following approaches are tested here with spectrograms for the first time: Local Ternary Phase Quantization, Heterogeneous Auto-Similarities of Characteristics, and an ensemble of variants of Local Binary Pattern Histogram Fourier. Combining this set of descriptors greatly increases classification performance and markedly improves previous ensembles of texture descriptors used for describing a spectrogram. Moreover, a further improvement is obtained when the texture descriptors are combined with the acoustic features. SVM classifiers are used in the classification step, with final results computed using 10-fold cross-validation. For a fair comparison with other methods in the literature, the experiments are performed on a benchmark database composed of 46 bird species used for this classification task. The best accuracy rate obtained is about 94.5%. The MATLAB code we used is publicly available to other researchers for future comparisons, as well as the database used in the experiments.

This paper describes a system for automatic bird species classification based on features taken from the textural content of spectrogram images. The texture features are extracted using three of the most common texture operators described in the Digital Image Processing literature: Local Binary Pattern (LBP), Local Phase Quantization (LPQ) and Gabor Filters. Aiming to perform more fare comparisons, the experiments were performed over a database already used in other works presented in the literature. In the classification step, SVM classifier was used and the final results were taken using 10-fold cross validation. The experiments were performed over a challenger dataset composed of 46 classes, and the best accuracy rate obtained is about 77.65%.

In this paper we compare two different textural feature sets for automatic music genre classification. The idea is to convert the audio signal into spectrograms and then extract features from this visual representation. Two textural descriptors are explored in this work: the Gray Level Co-Occurrence Matrix (GLCM) and Local Binary Patterns (LBP). Besides, two different strategies of extracting features are considered: a global approach where the features are extracted from the entire spectrogram image and then classified by a single classifier; a local approach where the spectrogram image is split into several zones which are classified independently and final decision is then obtained by combining all the partial results. The database used in our experiments was the Latin Music Database, which contains music pieces categorized into 10 musical genres, and has been used for MIREX (Music Information Retrieval Evaluation eXchange) competitions. After a comprehensive series of experiments we show that the SVM classifier trained with LBP is able to achieve a recognition rate of 80%. This rate not only outperforms the GLCM by a fair margin but also is slightly better than the results reported in the literature.