Face detection and its applications in intelligent and focused image retrieval

In this paper, a three-dimensional anisotropic diffusion equation is used to conduct an in-depth study and analysis of students’ concentration in video recognition in English teaching classrooms. A multifeature fusion face live detection method based on diffusion model extracts Diffusion Kernel (DK) features and depth features from diffusion-processed face images, respectively. DK features provide a nonlinear description of the correlation between successive face images and express face image sequences in the temporal dimension; depth features are extracted by a pretrained depth neural network model that can express the complex nonlinear mapping relationships of images and reflect the more abstract implicit information inside face images. To improve the effectiveness of the face image features, the extracted DK features and depth features are fused using a multicore learning method to obtain the best combination and the corresponding weights. The two features complement each other, and the fused features are more discriminative, which provides a strong basis for the live determination of face images. Experiments show that the method has excellent performance and can effectively discriminate the live nature of faces in images and resist forged face attacks. Based on the above face detection and expression recognition algorithms, the classroom concentration analysis system based on expression recognition is designed to achieve real-time acquisition and processing of classroom images, complete student classroom attendance records using face detection and face recognition methods, and analyze students’ concentration from the face integrity and facial expression of students facing the blackboard by combining face detection and expression recognition to visualize and display students’ classroom data for teachers, students, and parents with more data support and help.

Face and eye detection is one of the most challenging problems in computer vision area. The goal of this paper is to present a study over the existing literature on face and eye detection and gaze estimation. With the uptrend of systems based on face and eye detection in many different areas of life in recent years, this subject has gained much more attention by academic and industrial area. Many different studies have been performed about face and eye detection. Besides having many challenging problems like, having different lighting conditions, having glasses, facial hair or mustache on face, different orientation pose or occlusion of face, face and eye detection methods performed great progress. In this paper we first categorize face detection models and examine the basic algorithms for face detection. Then we present methods for eye detection and gaze estimation.

Fatigue driving detection helps reduce the occurrence of traffic accidents. Currently, the fatigue detection methods still have following challenges: 1) In the visual fatigue detection system, face detection is the foundation of fatigue detection and it is always susceptible to changes of illumination and facial expression, which lowers the accuracy of face detection. 2) It is so difficult to distinguish some behaviors of a driver such as singing and yawning in a single frame that making false fatigue detection in the visual fatigue detection system. To overcome the problems mentioned above, a fatigue detection method based on sequential mouth feature with long short-term memory network is proposed. The proposed method mainly consists of two parts: locating of facial landmark based on face and mouth detection, and yawning detection based on residual network (Resnet) and long short-term memory (LSTM). The face detection method is based on Histogram of Oriented Gradient (HOG) feature and Support Vector Machine (SVM) classifier, which is robust to the changes of illumination and slightly facial expression. The mouth detection is based on cascaded regression tree, which extracts the mouth region from the detected face. The Resnet and LSTM based yawning detection extracts spatial-temporal features from the mouth regions between frames and distinguishes the behaviors about yawning to identify fatigue of a driver with sequential mouth feature. The proposed method achieves accuracy of 94.9% on the YawDD dataset, which outperforms the compared methods. Experimental results demonstrate that the proposed method satisfies the performance of fatigue detection.

A conventional color histogram (CCH) considers neither the color similarity across different bins nor the color dissimilarity in the same bin. Therefore, it is sensitive to noisy interference such as illumination changes and quantization errors. Furthermore, CCHs large dimension or histogram bins requires large computation on histogram comparison. To address these concerns, this paper presents a new color histogram representation, called fuzzy color histogram (FCH), by considering the color similarity of each pixel's color associated to all the histogram bins through fuzzy-set membership function. A novel and fast approach for computing the membership values based on fuzzy c-means algorithm is introduced. The proposed FCH is further exploited in the application of image indexing and retrieval. Experimental results clearly show that FCH yields better retrieval results than CCH. Such computing methodology is fairly desirable for image retrieval over large image databases.

This paper presents a novel face detection method by applying distribution-based distance (DBD) measure and support vector machine (SVM). The novelty of our DBD-SVM method comes from the integration of discriminating feature analysis, face class modeling, and support vector machine for face detection. First, the discriminating feature vector is defined by combining the input image, its 1-D Haar wavelet representation, and its amplitude projections. Then the DBD-SVM method statistically models the face class by applying the discriminating feature vectors and defines the distribution-based distance measure. Finally, based on DBD and SVM, three classification rules are applied to separate faces and nonfaces. Experiments using images from the MIT-CMU test sets show the feasibility of our new face detection method. In particular, when using 92 images (containing 282 faces) from the MIT-CMU test sets, our DBD-SVM method achieves 98.2% correct face detection accuracy with 2 false detections, a performance comparable to the state-of-the-art face detection methods, such as the Schneiderman-Kanade's method.

We present a facial feature detector for time-of-flight (TOF) cameras that extends previous work by combining a nose detector based on geometric features with a face detector. The goal is to prevent false detections outside the area of the face. To detect the nose in the image, we first compute the geometric features per pixel. We then augment these geometric features with two additional features: The horizontal and vertical distance to the most likely face detected by a cascade-of-boosted-ensembles face detector. We use a very simple classifier based on an axis-aligned bounding box in feature space; pixels whose feature values fall within the box are classified as nose pixels, and all other pixels are classified as ldquonon-noserdquo. The extent of the bounding box is learned on a labeled training set. Despite its simiplicity, this detector already delivers satisfactory results on the geometric features alone; adding the face detector improves the equal error rate (EER) from 22.2% (without face detector) to 10.4% (with face detector). (Note when comparing with our previous results from [1] and [2] that, in contrast to this paper, the test data used there did not contain scale variations.).

This paper introduces the technique of graph cut into the extraction of object region and applies the corresponding result of object region extraction into the image retrieval based on object region. The main idea of image retrieval based on object region is to use the feature of object region instead of the feature of global image to participate in the image retrieval. In the field of graphics there is a technique called graph cut, which can be used to figure out the contour of object under the interaction of users. The graph cut algorithm can be used to verify the correctness of object region extraction, and the users' input about seeds can be simulated according to the initial object region extracted. The usage of graph cut can make the object region extracted more precisely and thus the performance of image retrieval based on object region can be improved. Experiments show that the object region extraction algorithm based on graph cut is valid and the subsequent image retrieval results accord with the human visual perception much more than the ones without the usage of graph cut.

This dissertation explores original techniques for the construction of hypergraph models for computer vision applications. A hypergraph is a generalization of a pairwise simple graph, where an edge can connect any number of vertices. The expressive power of the hypergraph models places a special emphasis on the relationship among three or more objects, which has made hypergraphs better models of choice in a lot of problems. This is in sharp contrast with the more conventional graph representation of visual patterns where only pairwise connectivity between objects is described. The contribution of this thesis is fourfold: (i) For the first time the advantage of the hypergraph neighborhood structure is analyzed. We argue that the summarized local grouping information contained in hypergraphs causes an ‘averaging’ effect which is beneficial to the clustering problems, just as local image smoothing may be beneficial to the image segmentation task. (ii) We discuss how to build hypergraph incidence structures and how to solve the related unsupervised and semi-supervised problems for three different computer vision scenarios: video object segmentation, unsupervised image categorization and image retrieval. We compare our algorithms with state-of-the-art methods and the effectiveness of the proposed methods is demonstrated by extensive experimentation on various datasets. (iii) For the application of image retrieval, we propose a novel hypergraph model — probabilistic hypergraph to exploit the structure of the data manifold by considering not only the local grouping information, but also the similarities between vertices in hyperedges. (iv) In all three applications mentioned above, we conduct an in depth comparison between simple graph and hypergraph based algorithms, which is also beneficial to other computer vision applications.

Content based image retrieval (CBIR) has become a research hotspot in the field of image processing because of its high theoretical and application value. Extracting and matching image features is the main method of CBIR. However, it is extremely difficult to extract the effective features of images. Based on HSV color space characteristics and human perception of color, a color recognition method is proposed. And the structure of this class of pixels is extracted by the method of classification. Image feature matching will be carried out among the same pixel sets, which reduces the complexity of image feature extraction and matching. Experimental results show that the proposed image retrieval method has a good effect.KeywordsImage retrievalFuzzy setColor recognitionFeature extraction

In this paper we propose a new distance metric for probability density functions (PDF). The main advantage of this metric is that unlike the popular Kullback-Liebler (KL) divergence it can be computed in closed form when the PDFs are modeled as Gaussian Mixtures (GM). The application in mind for this metric is histogram based image retrieval. We experimentally show that in an image retrieval scenario the proposed metric provides as good results as the KL divergence at a fraction of the computational cost. This metric is also compared to a Bhattacharyya-based distance metric that can be computed in closed form for GMs and is found to produce better results.

Face detection places an important role in face recognition which is a popular choice for biometric systems. To solve the low face detection rate problem of face detection in constrained scenario, an efficient face detection method based on migration learning was proposed in this paper. In the proposed facial detection approach, data-cleaning was firstly used to optimize the face database. Then the Visual Geometry Group 16 (VGG16) deep learning network was improved to realize migration learning by replacing the softmax regression layer with the multi-scale feature detection layer. Finally, the constrained scene face images for testing were detected and labeled by the trained migration learning model. The WIDER FACE dataset was used for experiments. Experiment results showed that the proposed method can successfully perform face detection in the WIDER FACE dataset and obtain more than 90% detection rate.

Since there are few open image retrieval toolkits available, researchers in the field are often forced to re-implement existing algorithms in order to perform a comparative evaluation. None of the existing toolkits support retrieval of JPEG images directly in the compressed domain. The authors’ aim is therefore to facilitate the use of compressed domain image retrieval techniques as well as ease retrieval evaluation by fellow researchers. For this purpose, the authors present JIRL, an open source C++ software suite that allows content-based image retrieval in the JPEG compressed domain and provides tools for benchmarking retrieval accuracy and retrieval time. In total, twelve state-of-the-art JPEG retrieval algorithms are implemented, while for each method techniques for compressed domain feature extraction as well as feature comparison are provided in an object-oriented framework. An example image retrieval application is also provided to demonstrate how the library can be used. JIRL is made available to fellow researchers under the LGPL v.2.1 license.

Achieving a balance between security and retrieval accuracy presents a significant challenge in secure content-based image retrieval (SCBIR), particularly in untrusted cloud environments. Conventional SCBIR schemes often struggle with trade-offs, either compromising data privacy, reducing retrieval performance, or introducing significant computational overhead, making them unsuitable for large-scale real-world applications. To overcome these challenges, we propose SCBIR-PE, a novel SCBIR scheme that integrates block-based perceptual encryption with an efficient histogram-based feature extraction method. The encryption incorporates geometric and color transformations at block and sub-block levels across three distinct extensions, confirming security robustness while preserving image features necessary for an image retrieval application. Thus, feature extraction is performed using the Color and Edge Directivity Descriptor (CEDD), enabling direct feature computation in the encrypted domain. Additionally, Euclidean distance is utilized as a similarity measurement technique, ensuring efficient image retrieval while avoiding computationally intensive tasks like clustering, index generation, or feature mapping, which are at the core of existing SCBIR schemes. Experimental evaluations demonstrate that SCBIR-PE scheme outperforms conventional privacy preserving image retrieval schemes in terms of retrieval accuracy (up to 4% better mean average precision score), security robustness, and computational efficiency. By ensuring strong privacy protection with minimal computational overhead, our proposed scheme offers a scalable and efficient solution for cloud-assisted secure image retrieval applications.

Content-based satellite cloud image retrieval is a very important problem in image processing and analysis field. Traditional image retrieval method has some limitation, for realized image retrieval accurately and quickly, the CBIR method is an adaptive method. For achieved good retrieval result, some of the pretreatment method of the satellite cloud image was used, and the experiment effect was shown. The basic theory and structure model of satellite cloud image database was studied systematically, and the data model of satellite cloud image database was provided. On the basis of studied the key technology of CBIR, we could obtain the better retrieval result, and the image retrieval result was shown in detail. At last, the application of CBIR method in satellite cloud image has been studied. The experiment result proves that the research and application of content-based satellite cloud image retrieval is valuable, which could improve the image search efficiency more.

Nowadays, face detection is common. Its used in many areas. With the help of face detection benchmark data-sets, many signs of progress have been made. Face detection methods used nowadays is not matching the real-world requirements. With new advancing technologies and services, we need to upgrade our existing systems. By using a data-set called as WIDER FACE which is very large in size than already existing data-sets, we can improve the performance. WIDER FACE data-set has many faces in it which may be challenging as it includes faces under different conditions. Moreover, we can see that in face detection task, WIDER FACE data-set is best for training and testing the accuracy of the model. But existing face detection algorithms and models are not up-to the mark. They have major limitations. So we created a WIDER FACE detection system which will help us overcome all those issues.