Search

In an object detection system, the main objective during training is to maintain the detection and false positive rates under acceptable levels when the model is run over the test set. However, this typically translates into an unacceptable rate of false alarms when the system is deployed in a real surveillance scenario. To deal with this situation, which often leads to system shutdown, we propose to add a filter step to discard part of the new false positive detections that are typical of the new scenario. This step consists of a deep autoencoder trained with the false alarm detections generated after running the detector over a period of time in the new scenario. Therefore, this step will be in charge of determining whether the detection is a typical false alarm of that scenario or whether it is something anomalous for the autoencoder and, therefore, a true detection. In order to decide whether a detection must be filtered, three different approaches have been tested. The first one uses the autoencoder reconstruction error measured with the mean squared error to make the decision. The other two use the k-NN (k-nearest neighbors) and one-class SVMs (support vector machines) classifiers trained with the autoencoder vector representation. In addition, a synthetic scenario has been generated with Unreal Engine 4 to test the proposed methods in addition to a dataset with real images. The results obtained show a reduction in the number of false positives between 22.5% and 87.2% and an increase in the system’s precision of 1.2%$$-47$$ - 47 % when the autoencoder is applied.

Introducing efficient automatic violence detection in video surveillance or audiovisual content monitoring systems would greatly facilitate the work of closed-circuit television (CCTV) operators, rating agencies or those in charge of monitoring social network content. In this paper we present a new deep learning architecture, using an adapted version of DenseNet for three dimensions, a multi-head self-attention layer and a bidirectional convolutional long short-term memory (LSTM) module, that allows encoding relevant spatio-temporal features, to determine whether a video is violent or not. Furthermore, an ablation study of the input frames, comparing dense optical flow and adjacent frames subtraction and the influence of the attention layer is carried out, showing that the combination of optical flow and the attention mechanism improves results up to 4.4%. The conducted experiments using four of the most widely used datasets for this problem, matching or exceeding in some cases the results of the state of the art, reducing the number of network parameters needed (4.5 millions), and increasing its efficiency in test accuracy (from 95.6% on the most complex dataset to 100% on the simplest one) and inference time (less than 0.3 s for the longest clips). Finally, to check if the generated model is able to generalize violence, a cross-dataset analysis is performed, which shows the complexity of this approach: using three datasets to train and testing on the remaining one the accuracy drops in the worst case to 70.08% and in the best case to 81.51%, which points to future work oriented towards anomaly detection in new datasets. Ministerio de Economía y Competitividad TIN2017-82113-C2-1-R MInisterio de Economía y Competitividad TIN2017-82113-C2-2-R

The data presented in this article is part of the whole slide imaging (WSI) datasets generated in European project AIDPATH 2 This data is also related to the research paper entitle “Glomerulosclerosis Identification in Whole Slide Images using Semantic Segmentation”, published in Computer Methods and Programs in Biomedicine Journal [1]. In that article, different methods based on deep learning for glomeruli segmentation and their classification into normal and sclerotic glomerulous are presented and discussed. The raw data used is described and provided here. In addition, the detected glomeruli are also provided as individual image files. These data will encourage research on artificial intelligence (AI) methods, create and compare fresh algorithms, and measure their usability in quantitative nephropathology. Keywords: Glomeruli identification, Normal glomerulus, Global sclerotic glomerulus, Whole slide image, Digital pathology

This chapter presents the most relevant image processing techniques and algorithms related to computing features that are able to characterize diatoms as objects in the computer vision field for further analysis and classification. For this purpose, a wide revision of the most important contributions to diatom classification is performed. Moreover, features that have been found to be suitable for this task are covered. Later on, the reader will find the main techniques for diatom classification for the two paradigms that are used nowadays: machine learning with classical methods that rely on previously selected features, or deep learning, which learns the features from the images automatically.

15 pags., 21 figs., 4 tabs., Microscopic algae segmentation, specifically of diatoms, is an essential procedure for water quality assessment. The segmentation of these microalgae is still a challenge for computer vision. This paper addresses for the first time this problem using deep learning approaches to predict exactly those pixels that belong to each class, i.e., diatom and non diatom. A comparison between semantic segmentation and instance segmentation is carried out, and the performance of these methods is evaluated in the presence of different types of noise. The trained models are then evaluated with the same raw images used for manual diatom identification. A total of 126 images of the entire field of view at 60x magnification, with a size of 2592x1944 pixels, are analyzed. The images contain 10 different taxa plus debris and fragments. The best results were obtained with instance segmentation achieving an average precision of 85% with 86% sensitivity and 91% specificity (up to 92% precision with 98%, both sensitivity and specificity for some taxa). Semantic segmentation was able to improve the average sensitivity up to 95% but decreasing the specificity down to 60% and precision to 57%. Instance segmentation was also able to properly separate diatoms when overlap occurs, which helps estimate the number of diatoms, a key requirement for water quality grading., The authors received financial support from the Spanish Government under the Aqualitas-retos project (Ref. CTM2014-51907-C2-2-R-MINECO).

IbPRIA 2019,Madrid, Spain. July 1-4 2019. -- http://www.ibpria.org/2019/. -- Part of the Lecture Notes in Computer Science book series (LNCS, volume 11867) Also part of the Image Processing, Computer Vision, Pattern Recognition, and Graphics book sub series (LNIP, volume 11867), Diatom identification is a crucial process to estimate water quality, which is essential in biological studies. This process can be auto- mated with machine learning algorithms. For this purpose, a dataset with 10 common taxa is collected, with annotations provided by an expert diatomist. In this work, a comparison of the classical state-of-the-art gen- eral purpose methods along with two different deep learning approaches is carried out. The classical methods are based on Viola-Jones and scale and curvature invariant ridge object detectors. The deep learning based methods are Semantic Segmentation and YOLO. This is the first time that Viola-Jones and Semantic Segmentation techniques are applied and compared for diatom segmentation in microscopic images containing sev- eral taxon shells. While all methods provide relatively good results in specific species, the deep learning approaches are consistently better in terms of sensitivity and specificity (up to 0.99 for some taxa) and up to 0.86 precision., The authors acknowledge financial support of the Spanish Government under the Aqualitas-retos project (Ref. CTM2014-51907-C2-R-MINECO).

The development of object detection systems is normally driven to achieve both high detection and low false positive rates in a certain public dataset. However, when put into a real scenario the result is generally an unacceptable rate of false alarms. In this context we propose to add an additional step that models and filters the typical false alarms of the new scenario while roughly maintaining the ability to detect the objects of interest. We propose to use the false alarms of the new scenario to train a deep autoencoder and to model them. The latter will act as a filter that checks whether the output of the detector is one of its typical false positives or not based on the reconstruction error measured with the Mean Squared Error (MSE) and the Peak Signal-to-Noise Ratio (PSNR). We test the system using an entirely synthetic novel dataset for training and testing the autoencoder generated with Unreal Engine 4. Results show a reduction in the number of FPs of up to 37.9% in combination with the PSNR error while maintaining the same detection capability.

Background and Objective: Glomeruli identification, i.e., detection and characterization, is a key procedure in many nephropathology studies. In this paper, semantic segmentation based on convolutional neural networks (CNN) is proposed to detect glomeruli using Whole Slide Imaging (WSI) follows by a classification CNN to divide the glomeruli into normal and sclerosed. Methods: Comparison between U-Net and SegNet CNNs is performed for pixel-level segmentation considering both a two and three class problem, that is, a) non-glomerular and glomerular structures and b) non-glomerular normal glomerular and sclerotic structures. The two class semantic segmentation result is then used for a CNN classification where glomerular regions are divided into normal and global sclerosed glomeruli. Results: These methods were tested on a dataset composed of 47 WSIs belonging to human kidney sections stained with Periodic Acid Schiff (PAS). The best approach was the SegNet for two class segmentation follows by a fine-tuned AlexNet network to characterize the glomeruli. 98.16% of accuracy was obtained with this process of consecutive CNNs (SegNet-AlexNet) for segmentation and classification. Conclusion: The results obtained demonstrate that the sequential CNN segmentation-classification strategy achieves higher accuracy reducing misclassified cases and therefore being the methodology proposed for glomerulosclerosis detection.

The goal of this challenge 1 was to evaluate new and existing algorithms for automated detection of metastases in hematoxylin and eosin (H&E) stained whole-slide images of lymph node sections. To this end, both slide-based and lesion-based evaluation was made. Several classification methods were tested including classical and deep learning techniques. The most efficient one for the dataset tested was Bagging Tree classifiers using texture features. In the slide-based classification an AUC equal to 0.9952 was obtained, with 98.13% of TP and 1.28% of FP. The TP result decreases in the lesion-based evaluation. Two methodologies were proposed for this second evaluation. Method 1 was based on the convex area of the regions and method 2 based on morphophone- mic, geometric and statistical features. The sensitivity for lesion detection was 39.83% and 36.66% respectively, though the false positive average is kept low about 12.28 in method 1 and 10.71 in method 2.

Advances in digital pathology are generating huge volumes of whole slide (WSI) and tissue microarray images (TMA) which are providing new insights into the causes of cancer. The challenge is to extract and process effectively all the information in order to characterize all the heterogeneous tissue-derived data. This study aims to identify an optimal set of features that best separates different classes in breast TMA. These classes are: stroma, adipose tissue, benign and benign anomalous structures and ductal and lobular carcinomas. To this end, we propose an exhaustive assessment on the utility of textons and colour for automatic classification of breast TMA. Frequential and spatial texton maps from eight different colour models were extracted and compared. Then, in a novel way, the TMA is characterized by the 1st and 2nd order Haralick statistical descriptors obtained from the texton maps with a total of 241 × 8 features for each original RGB image. Subsequently, a feature selection process is performed to remove redundant information and therefore to reduce the dimensionality of the feature vector. Three methods were evaluated: linear discriminant analysis, correlation and sequential forward search. Finally, an extended bank of classifiers composed of six techniques was compared, but only three of them could significantly improve accuracy rates: Fisher, Bagging Trees and AdaBoost. Our results reveal that the combination of different colour models applied to spatial texton maps provides the most efficient representation of the breast TMA. Specifically, we found that the best colour model combination is Hb, Luv and SCT for all classifiers and the classifier that performs best for all colour model combinations is the AdaBoost. On a database comprising 628 TMA images, classification yields an accuracy of 98.1% and a precision of 96.2% with a total of 316 features on spatial textons maps.

Face recognition has been a long standing problem in computer vision. Recently, Histograms of Oriented Gradients (HOGs) have proven to be an effective descriptor for object recognition in general and face recognition in particular. In this paper, we investigate a simple but powerful approach to make robust use of HOG features for face recognition. The three main contributions of this work are: First, in order to compensate for errors in facial feature detection due to occlusions, pose and illumination changes, we propose to extract HOG descriptors from a regular grid. Second, fusion of HOG descriptors at different scales allows to capture important structure for face recognition. Third, we identify the necessity of performing dimensionality reduction to remove noise and make the classification process less prone to overfitting. This is particularly important if HOG features are extracted from overlapping cells. Finally, experimental results on four databases illustrate the benefits of our approach.

Automatic detection systems usually require large and representative training datasets in order to obtain good detection and false positive rates. Training datasets are such that the positive set has few samples and/or the negative set should represent anything except the object of interest. In this respect, the negative set typically contains orders of magnitude more images than the positive set. However, imbalanced training databases lead to biased classifiers. In this paper, we focus our attention on a negative sample selection method to properly balance the training data for cascade detectors. The method is based on the selection of the most informative false positive samples generated in one stage to feed the next stage. The results show that the proposed cascade detector with sample selection obtains on average better partial AUC and smaller standard deviation than the other compared cascade detectors.

Support vector machines (SVM) and independent component analysis (ICA) are two powerful and relatively recent techniques. SVMs are classifiers which have demonstrated high generalization capabilities in many different tasks, including the object recognition problem. ICA is a feature extraction technique which can be considered a generalization of principal component analysis (PCA). ICA has been mainly used on the problem of blind signal separation. In this paper we combine these two techniques for the face recognition problem. Experiments were made on two different face databases, achieving very high recognition rates. As the results using the combination PCA/SVM were not very far from those obtained with ICA/SVM, our experiments suggest that SVMs are relatively insensitive to the representation space. Thus as the training time for ICA is much larger than that of PCA, this result indicates that the best practical combination is PCA with SVM.

Breast cancer diagnosis is still done by observation of biopsies under the microscope. The development of automated methods for breast TMA classification would reduce diagnostic time. This paper is a step towards the solution for this problem and shows a complete study of breast TMA classification based on colour models and texture descriptors. The TMA images were divided into four classes: i) benign stromal tissue with cellularity, ii) adipose tissue, iii) benign and benign anomalous structures, and iv) ductal and lobular carcinomas. A relevant set of features was obtained on eight different colour models from first and second order Haralick statistical descriptors obtained from the intensity image, Fourier, Wavelets, Multiresolution Gabor, M-LBP and textons descriptors. Furthermore, four types of classification experiments were performed using six different classifiers: (1) classification per colour model individually, (2) classification by combination of colour models, (3) classification by combination of colour models and descriptors, and (4) classification by combination of colour models and descriptors with a previous feature set reduction. The best result shows an average of 99.05% accuracy and 98.34% positive predictive value. These results have been obtained by means of a bagging tree classifier with combination of six colour models and the use of 1719 non-correlated (correlation threshold of 97%) textural features based on Statistical, M-LBP, Gabor and Spatial textons descriptors.

This paper deals with automatic taxa identification based on machine learning methods. The aim is therefore to automatically classify diatoms, in terms of pattern recognition terminology. Diatoms are a kind of algae microorganism with high biodiversity at the species level, which are useful for water quality assessment. The most relevant features for diatom description and classification have been selected using an extensive dataset of 80 taxa with a minimum of 100 samples/taxon augmented to 300 samples/taxon. In addition to published morphological, statistical and textural descriptors, a new textural descriptor, Local Binary Patterns (LBP), to characterize the diatom’s valves, and a log Gabor implementation not tested before for this purpose are introduced in this paper. Results show an overall accuracy of 98.11% using bagging decision trees and combinations of descriptors. Finally, some phycological features of diatoms that are still difficult to integrate in computer systems are discussed for future work., The authors acknowledge financial support of the Spanish Government under the Aqualitas-retos project (Ref. CTM2014-51907-C2-2-R-MINECO), We acknowledge support by the CSIC Open Access Publication Initiative through its Unit of Information Resources for Research (URICI).

This paper shows a complete and novel study for breast TMA classification based on texture, frequential features and color models information. Thus a relevant set of features for automatic breast TMA classification is found. These features are obtained from 1st and 2nd order Haralick statistical descriptors as well as by filtering with the Fourier and Daubechies Wavelet transforms. Moreover, the discriminant value of the features has been analyzed under different color models, that is, RGB, Lab*, HSV, Hb* and b* channel. Thus, a total of 3133 features with 5 color models were analyzed. The TMA images were divided into four classes: i) benign stromal tissue with cellularity, ii) adipose tissue, iii) benign structures but anomalous and iv) ductal and lobular carcinomas. A statistical study of the features was carried out to provide information about their influence in the tissue classes. Finally, the classification was performed with 10 classifiers using 10-fold and leave-one-out cross validation for all color models selected. Furthermore, the classification results were also analyzed by means of a principal component analysis (PCA). A 95% accuracy and 90% precision with PCA and a Fisher classifier on Hb* image channels is obtained with the selected features.

42 pags. ; 5 figs. ; 3 tabs., Pollen identification is required in different scenarios such as prevention of allergic reactions, climate analysis or apiculture. However, it is a time-consuming task since experts are required to recognize each pollen grain through the microscope. In this study, we performed an exhaustive assessment on the utility of texture analysis for automated characterization of pollen samples. A database composed of 1800 brightfield microscopy images of pollen grains from 15 different taxa was used for this purpose. A pattern recognition-based methodology was adopted to perform pollen classification. Four different methods were evaluated for texture feature extraction from the pollen image: Haralick's gray-level co-occurrence matrices (GLCM), log-Gabor filters (LGF), local binary patterns (LBP) and discrete Tchebichef moments (DTM). Fisher's discriminant analysis and k-nearest neighbour were subsequently applied to perform dimensionality reduction and multivariate classification, respectively. Our results reveal that LGF and DTM, which are based on the spectral properties of the image, outperformed GLCM and LBP in the proposed classification problem. Furthermore, we found that the combination of all the texture features resulted in the highest performance, yielding an accuracy of 94.83%. Therefore, thorough texture characterization could be considered in further implementations of automatic pollen recognition systems based on image processing techniques., J. V. Marcos is a research fellow at Institute of Optics (CSIC) under the programme Juan de la Cierva (Spanish Ministry of Economy and Competitiveness). R. Nava thanks Consejo Nacional de Ciencia y Tecnologa (CONACYT) and PAPIIT grant IG100814.

This chapter presents a comparative study among different colour models (RGB, HSI, CMYK, CIEL*a*b*, and HSD) applied to very large microscopic image analysis. Such analysis of different colour models is needed in order to carry out a successful detection and therefore a classification of different regions of interest (ROIs) within the image. This, in turn, allows both distinguishing possible ROIs and retrieving their proper colour for further ROI analysis. This analysis is not commonly done in many biomedical applications that deal with colour images. Other important aspect is the computational cost of the different processing algorithms according to the colour model. This work takes these aspects into consideration to choose the best colour model tailored to the microscopic stain and tissue type under consideration and to obtain a successful processing of the histological image.

[Abstract] This paper describes a novel weighted voting tree classification scheme for breast density classification. Breast parenchymal density is an important risk factor in breast cancer. Moreover, it is known that mammogram interpretation is more difficult when dense tissue is involved. Therefore, automated breast density classification may aid in breast lesion detection and analysis. Several classification methods have been compared and a novel hierarchical classification procedure of combined classifiers with linear discriminant analysis (LDA) is proposed as the best solution to classify the mammograms into the four BIRADS tissue classes. The classification scheme is based on 298 texture features. Statistical analysis to test the normality and homoscedasticity of the data was carried out for feature selection. Thus, only features that are influenced by the tissue type were considered. The novel classification techniques have been incorporated into a CADe system to drive the detection algorithms and tested with 1459 images. The results obtained on the 322 screen-film mammograms (SFM) of the mini-MIAS dataset show that 99.75% of samples were correctly classified. On the 1137 full-field digital mammograms (FFDM) dataset results show 91.58% agreement. The results of the lesion detection algorithms were obtained from modules integrated within the CADe system developed by the authors and show that using breast tissue classification prior to lesion detection leads to an improvement of the detection results. The tools enhance the detectability of lesions and they are able to distinguish their local attenuation without local tissue density constraints.

The development of a detection system is normally driven to achieve good detection rates. In most cases, a good detection rate involves a number of false positive decisions. However, the false positive rate is ultimately what decides if the detection system is effective or not. Another aspect to consider in automatic detection systems is the time to analyse an image until a decision is made. Viola & Jones proposed a cascade detector that achieves good detection and false positive rates at high speed. Some authors have proposed modifications to the cascade detector in order to improve the detection rate while maintaining the same false positive rate. However, during the implantation of the system we consistently find a large number of false positive detections due to the lack of knowledge about the newly acquired images. In this work, we propose a parallel cascade detector that gradually incorporates these new false positives to achieve an acceptable false positive rate. The second cascade detector is built using the new false positive detection images and the original true positive images during the implantation period. The proposed parallel scheme reduces the false positive rate of the system at roughly the same speed.

Given that angiogenesis and lymphangiogenesis are strongly related to prognosis in neoplastic and other pathologies and that many methods exist that provide different results, we aim to construct a morphometric tool allowing us to measure different aspects of the shape and size of vascular vessels in a complete and accurate way. The developed tool presented is based on vessel closing which is an essential property to properly characterize the size and the shape of vascular and lymphatic vessels. The method is fast and accurate improving existing tools for angiogenesis analysis. The tool also improves the accuracy of vascular density measurements, since the set of endothelial cells forming a vessel is considered as a single object.

This paper describes a specific tool to automatically perform the segmentation and archiving of tissue microarray (TMA) cores in microscopy images at different magnification, that is, 5x, 10x, 20x and 40x. TMA enables researchers to extract small cylinder of single tissues (core sections) from histological sections and arrange them in an array on a paraffin block such that hundreds can be analyzed simultaneously. A crucial step to improve the speed and quality of these analyses is the correct recognition of each tissue position in the array. However, usually the tissue cores are not aligned in the microarray, the TMA cores are broken and the digital images are noisy. We develop a robust framework to handle core sections under these conditions. The algorithms are able to detect, stitch and archive the TMA cores. Once the TMA cores are segmented they are stored in a relational database allowing their location and classification for further studies of benign-malignant classification. The method was shown to be reliable for handling the TMA cores and therefore enabling further large scale molecular pathology investigations.

The human face provides useful information during interaction; therefore, any system integrating Vision-Based Human Computer Interaction requires fast and reliable face and facial feature detection. Different approaches have focused on this ability but only open source implementations have been extensively used by researchers. A good example is the Viola–Jones object detection framework that particularly in the context of facial processing has been frequently used. The OpenCV community shares a collection of public domain classifiers for the face detection scenario. However, these classifiers have been trained in different conditions and with different data but rarely tested on the same datasets. In this paper, we try to fill that gap by analyzing the individual performance of all those public classifiers presenting their pros and cons with the aim of defining a baseline for other approaches. Solid comparisons will also help researchers to choose a specific classifier for their particular scenario. The experimental setup also describes some heuristics to increase the facial feature detection rate while reducing the face false detection rate.

Breast tissue density is an important risk factor in the detection of breast cancer It is also known that interpretation of mammogram lesions is more difficult in dense tissues Therefore, getting a preliminary tissue classification may aid in the subsequent process of breast lesion detection and analysis This article reviews several classification techniques for two datasets, both digitized screen-film (SFM) and full-field digital (FFDM) mammography, classified according to BIRADS categories It concludes with a hierarchical classification procedure based on k-NN combined with principal component analysis on texture features The results obtained classifying 1740 mammograms reflect up to 83% of samples correctly classified The method is being integrated within a CADe system developed by the authors.

Facial expression recognition has been the subject of much research in the last years within the Computer Vision community. The detection of smiles, however, has received less attention. Its distinctive configuration may pose less problem than other, at times subtle, expressions. On the other hand, smiles can still be very useful as a measure of happiness, enjoyment or even approval. Geometrical or local-based detection approaches like the use of lip edges may not be robust enough and thus researchers have focused on applying machine learning to appearance-based descriptors. This work makes an extensive experimental study of smile detection testing the Local Binary Patterns (LBP) as main descriptors of the image, along with the powerful Support Vector Machines classifier. The results show that error rates can be acceptable, although there is still room for improvement.

In this work local binary patterns based focus measures are presented. Local binary patterns (LBP) have been introduced in computer vision tasks like texture classification or face recognition. In applications where recognition is based on LBP, a computational saving can be achieved with the use of LBP in the focus measures. The behavior of the proposed measures is studied to test if they fulfill the properties of the focus measures and then a comparison with some well know focus measures is carried out in different scenarios.

In face recognition, where high-dimensional representation spaces are generally used, it is very important to take advantage of all the available information. In particular, many labelled facial images will be accumulated while the recognition system is functioning, and due to practical reasons some of them are often discarded. In this paper, we propose an algorithm for using this information. The algorithm has the fundamental characteristic of being incremental. On the other hand, the algorithm makes use of a combination of classification results for the images in the input sequence. Experiments with sequences obtained with a real person detection and tracking system allow us to analyze the performance of the algorithm, as well as its potential improvements.

Support Vector Machines (SVM) and Independent Component Analysis (ICA) are two powerful and relatively recent techniques. SVMs are classifiers which have demonstrated high generalization capabilities in many different tasks, including the object recognition problem. ICA is a feature extraction technique which can be considered a generalization of Principal Component Analysis (PCA). ICA has been mainly used on the problem of blind signal separation. In this paper we combine these two techniques for the face recognition problem. Experiments were made on two different face databases, achieving very high recognition rates. As the results using the combination PCA/SVM were not very far from those obtained with ICA/SVM, our experiments suggest that SVMs are relatively insensitive to the representation space. Thus as the training time for ICA is much larger than that of PCA, this result indicates that the best practical combination is PCA with SVM.