PYRAMID MATCH KERNEL AND CLASSIFIER ENSEMBLE-BASED MIL ALGORITHM FOR PORNOGRAPHIC IMAGES FILTERING

Aiming at the problem of object-based image retrieval, a novel semi-supervised multi-instance learning (MIL) algorithm based on RS (rough set) attribute reduction and Transductive support vector machine (TSVM) has been presented---RSTSVM- MIL algorithm. This algorithm regards the whole image as a bag, and the low-level visual feature of the segmented regions as instances, in order to transform every bag to a single sample, RSTSVM-MIL first generates a collection of visual-word to construct projection space by K-means clustering method in the instance feature space, then a nonlinear mapping is defined using these visual-word and maps every bag to a point in the projection space, extracting all bags' projection feature, which converts MIL problem to a standard supervised learning problem. Finally, in order to improve the efficiency of classifier training and to use a large amount of unlabeled image together with the labeled image to build better classifiers, we use RS (rough set) method to reduce the redundant information in projection feature firstly, then semi-supervised learning algorithm TSVM be used to train classifier. Experimental results on the SIVAL image set show that solving MIL problems using semi-supervised learning is feasible, and the RSTSVM-MIL is competitive with other state-of-the-art MIL algorithms in object- based image retrieval problem.

Multi-instance learning (MIL) has been widely applied to diverse applications involving complicated data objects, such as images and genes. However, most existing MIL algorithms can only handle small- or moderate-sized data. In order to deal with large-scale MIL problems, we propose MIL based on the vector of locally aggregated descriptors representation (miVLAD) and MIL based on the Fisher vector representation (miFV), two efficient and scalable MIL algorithms. They map the original MIL bags into new vector representations using their corresponding mapping functions. The new feature representations keep essential bag-level information, and at the same time lead to excellent MIL performances even when linear classifiers are used. Thanks to the low computational cost in the mapping step and the scalability of linear classifiers, miVLAD and miFV can handle large-scale MIL data efficiently and effectively. Experiments show that miVLAD and miFV not only achieve comparable accuracy rates with the state-of-the-art MIL algorithms, but also have hundreds of times faster speed. Moreover, we can regard the new miVLAD and miFV representations as multiview data, which improves the accuracy rates in most cases. In addition, our algorithms perform well even when they are used without parameter tuning (i.e., adopting the default parameters), which is convenient for practical MIL applications.

In this paper, the multi-instance learning algorithm is improved under the image retrieval framework based on contents, and the improved multi-instance learning algorithm is applied to image retrieval to better handle the ambiguity of the image. In this method, the image is used as the multi-instance bag and is divided into multiple instances by image segmentation algorithm, and then the multi-instance learning is performed with the multi-objective-diverse-density algorithm. The learning results are ordered by image similarity using the vector space model. Finally, relevant feedback is given in accordance with the positive bag and negative bag chosen by the user to provide satisfactory results to the user.

In this study, a novel algorithm for recognizing pornographic images based on the analysis of skin color regions is presented. The skin color information essentially provides Regions of Interest (ROIs). It is demonstrated that the convex hull of these ROIs provides semantically useful information for pornographic image detection. Based on these convex hulls, the authors extract a small set of low-level visual features that are empirically proven to possess discriminative power for pornographic image classification. In this study, the authors consider multi-class pornographic image classification, where the “nude” and “benign” image classes are further split into two specialized sub-classes, namely “bikini”/”porn” and “skin”/”non-skin”, respectively. The extracted feature vectors are fed to an ensemble of random forest classifiers for image classification. Each classifier is trained on a partition of the training set and solves a binary classification problem. In this sense, the model allows for seamless coarse-to-fine-grained classification by means of a tree-structured topology of a small number of intervening binary classifiers. The overall technique is evaluated on the AIIA-PID challenge of 9,000 samples of pornographic and benign images. The technique is shown to exhibit state-of-the-art performance against publicly available integrated pornographic image classifiers.

In multi-instance nonparallel plane learning (NPL), the training set is comprised of bags of instances and the nonparallel planes are trained to classify the bags. Most of the existing multi-instance NPL methods are proposed based on a twin support vector machine (TWSVM). Similar to TWSVM, they use only a single plane to generalize the data occurrence of one class and do not sufficiently consider the boundary information, which may lead to the limitation of their classification accuracy. In this article, we propose a multi-instance nonparallel tube learning (MINTL) method. Distinguished from the existing multi-instance NPL methods, MINTL embeds the boundary information into the classifier by learning a large-margin-based -tube for each class, such that the boundary information can be incorporated into refining the classifier and further improving the performance. Specifically, given a -class multi-instance dataset, MINTL seeks -tubes, one for each class. In multi-instance learning, each positive bag contains at least one positive instance. To build up the -tube of class , we require that each bag of class should have at least one instance included in the -tube. Moreover, except for one instance included in the -tube, the remaining instances in the positive bag may include positive instances or irrelevant instances, and their labels are unavailable. A large margin constraint is presented to assign the remaining instances either inside the -tube or outside the -tube with a large margin. Substantial experiments on real-world datasets have shown that MINTL obtains significantly better classification accuracy than the existing multi-instance NPL methods.

Multi-instance learning (MIL) has received more and more attentions in the machine learning research field due to its theoretical interest and its applicability to diverse real-world problems. In this paper, we present a probabilistic kernel approach for the multi-instance learning problems with various multi-instance assumptions by imposing Gaussian process prior on an unobservable latent function defined on the instance space. Because the relationship between the bag and its instances, triggered by the multi-instance assumption, can be exactly captured by defining the likelihood function, we can deal with different multi-instance assumptions by employing different likelihood functions. Experimental results on several multi-instance problems show that the proposed algorithms are valid and can achieve superior performance to the published MIL algorithms.

In this paper, a multi-instance learning based CBIR (content-based image retrieval) approach is presented, and multi-instance learning method is applied in CBIR, in order to deal with the inherent ambiguity of images. First of all the whole image is regards as a multi-instance bag, secondly the image is partitioned into multi-regions by using adaptive k-means image segmentation method, and then query images posed by the user are transformed into corresponding positive and negative bags and a EM-DD(expectation maximization diverse density) algorithm is employed for image retrieval and relevance feedback. Finally, it makes the users get satisfying result.

Infrared (IR) spectroscopy is a powerful tool for characterizing molecular structures and chemical groups, offering advantages such as low cost, rapid analysis, and non-destructive testing. When analyzing heterogeneous objects, spectra are typically measured from different regions to capture the local variations, presenting a multi-instance learning (MIL) problem. However, existing methods primarily rely on multi-instance assumptions or explicit bag representations, often failing to fully capture the intrinsic information from implicit representations. We introduce a bag dissimilarity regularized MIL framework for analyzing IR spectra of heterogeneous objects, which integrates both explicit and implicit representations to effectively learn the MIL bags. Specifically, a bag dissimilarity regularization term is utilized to extract implicit representations, which subsequently guide the classifier based on explicit representations to enhance generalization performance. The proposed method was validated on two heterogeneous detection tasks: polydimethylsiloxane (PDMS) block assessment and polyethylene terephthalate (PET) fiber inspection. Experimental results demonstrate that our approach significantly outperforms existing methods on both datasets with a considerable margin.

In real world, an image is usually associated with multiple labels which are characterized by different regions in the image. Thus image classification is naturally posed as both a multi-label learning and multi-instance learning problem. Different from existing research which has considered these two problems separately, we propose an integrated multi-label multi-instance learning (MLMIL) approach based on hidden conditional random fields (HCRFs), which simultaneously captures both the connections between semantic labels and regions, and the correlations among the labels in a single formulation. We apply this MLMIL framework to image classification and report superior performance compared to key existing approaches over the MSR Cambridge (MSRC) and Corel data sets.

Remotely sensed image grading advanced considerably; takes into account the availability and abundance of various resolution image grading algorithms. Several works were successful by the fusion of space-spectrum knowledge with supporting vector machines (SVM). To incorporate all these data with the composite approach, we suggest a technique using a hybrid multi-spectral and multi-instance procedure. This paper introduces a groundbreaking approach to exploring urban buildings through the implementation of the SVM-based support classification and Multi-instance learning (MIL). In this paper, we present the use of this model, the classification of images. Use high-resolution technology from Quickbird. This practice and archery have contributed to the performance, efficiency, and power. The suggested solution was tested in traditional urban imagery scenes. The results show a major improvement the classification performance compared to the two separately used attributes. The results of the experiments indicate a very promising accuracy of 91, 24%.

Learning in an image retrieval scheme that uses hidden annotations is investigated. Compared with low level visual features that are straightforward functions of the raw pixel values, hidden annotations are higher level hidden semantic attributes. In the proposed scheme, a small set of images is manually labeled with several hidden annotations. For each annotation, a support vector machine (SVM) classifier is trained using the images labeled with it as positive examples and others as negative examples. Based on the trained SVMs, the annotations are propagated to the unlabeled images in the database. To perform relevance feedback in the annotation space, a probabilistic re-weighting algorithm is proposed. Experimental results on a general-purpose database of 10,000 images demonstrate the potential of hidden annotation-based image retrieval and the superiority of the proposed relevance feedback algorithm over two existing algorithms.

Multi-instance learning (MIL) has many applications, including image and text categorization. One of the most effective approaches to MIL is by using Support Vector Machines with multi-instance kernels. In this paper we propose a multi-instance kernel, called MIR-Kernel, that takes into account the relational information of instances when computing similarities between bags. The relational information of instances are derived from the statistics of the distances between instances in feature space. The aim of MIR-Kernel is to efficiently capture the context in which instances occur within bags, so that it is able to better compute the similarities between bags. Experimental results on image and text categorization demonstrate the effectiveness of the proposed method compared to other methods.

In content-base image retrieval, relevance feedback (RF) schemes based on support vector machine (SVM) have been widely used to narrow the semantic gap between low-level visual features and high-level human perception. However, the performance of image retrieval with SVM active learning is known to be poor when the training data is insufficient. In this paper, the problem is solved by incorporating the unlabelled images into the learning process. We proposed a semi-supervised active learning algorithm which uses not only labeled training samples but also unlabeled ones to build better models. In relevance feedback, active learning algorithm is often used to reduce the cost of labeling by selecting only the most informative data. In addition, we introduced a semi-supervised approach which employed Nearest-Neighbor technique to label the unlabeled sample with a certain degree of uncertainty in its class information. Using these samples, Fuzzy support vector machine (FSVM) which takes into account the fuzzy nature of some training samples during its training is trained. We compared our method with standard active SVM on a database of 10,000 images, the experiment results show that the efficiency of SVM active learning can be improved by incorporating unlabelled images, and thus improve the overall retrieval performance. DOI: http://dx.doi.org/10.11591/telkomnika.v11i7.2807

The influence of architectural design on well-being is increasingly recognised, underscoring the need to understand how humans evaluate their surroundings. This study examines the impact of low-level visual features and high-level architectural elements in images of house facades generated by a generative adversarial network (GAN). We introduce a novel method to create controlled image datasets using the style mixing component of StyleGAN2-ADA, enabling the combination of specific architectural features with styles representing low-level image properties, such as brightness or colour. Our dataset consists of 900 images encompassing five high-level features (house size, garage door visibility, number of windows, entrance door visibility, and roof shape) and 16 low-level visual features. These GAN-generated images were rated in an online experiment by 303 participants on six dimensions: facelikeness, hominess, relaxation, invitingness, safety, and price. Linear mixed-effects models identified significant predictors, including the number of floors as a high-level feature and green pixel percentage, saturation (SD), brightness (SD), and contrast as low-level features. This indicated that houses with two floors and images with an overall higher amount of green pixel, as well as diversity of saturation and brightness and a lower contrast were rated as, e.g., more inviting and safer. The analysis showed that high- and low-level features explained up to 54% of the variance in house price ratings, with high-level features accounting for the larger share and highlighting their relevance for house facade evaluations. This work advances the use of GANs and offers a method to modify visual elements that shape architectural evaluations.

In this paper, we present an automatic web image and video mining framework with the ultimate goal of building a universal human age estimator based on facial information, which is applicable to all ethnic groups and various image qualities. On one hand, a large (391 k) yet noisy human aging image database is collected from Flickr and Google Image using a set of human age-related text queries. Multiple human face detectors based on distinctive techniques are adopted for noise-prune face detection. For each image, the detected faces with high detection confidences constitute a bag of face instances. We further remove the outliers via principal component analysis (PCA), which results in a condensed image database with about 175 k face instances. A robust multi-instance regressor learning algorithm is then developed to learn the kernel regression-based human age estimator in the presence of bag label noises. On the other hand, about 10 k video clips are downloaded from YouTube. We extract tracked face sequences from these video clips. Although their age labels are unknown, the tracked faces within a sequence are naturally with identical ages. This age-consistence constraint for face pairs is used as an extra regularizer to enhance the robustness of the age estimator. The derived human age estimator is extensively evaluated on three benchmark human aging databases, and without taking any images from these benchmark databases as training samples, comparable age estimation accuracies with the state-of-the-art results are achieved.