Photography is the activity of recording precious moments which are often difficult to make up afterwards. Therefore, taking the correct picture under proper guidance assistance is important. Although there are many factors that can determine a good photo, in general, photos that do not follow the composition rules usually look bad that make the viewer feel uncomfortable. Acting as a solution, in this paper, we propose an intelligent photographing guidance system using machine learning. The guidance is based on a tree-based interpretable machine learning model that can give reasons for decisions. There are two categories of features for guidance, which are traditional image features and deep features. Traditional features include prominent lines and image maps, such as saliency map and sharpness map, each of which exists in a multi-scale Gaussian pyramid. Deep features are extracted during the establishment of a CNN-based image composition classifier. We use these two categories of features as inputs for the interpretable machine learning model to establish a feasible photographing guidance system. The guidance system references our composition classifier with precision rate of 94.8%, and recall rate of 95.0% where the comprising tree-based interpretable model is capable of guiding camera users to alter image contents for obtaining better aesthetical compositions to take photos of good quality.

Photo aesthetics is a field in computer vision, which aims at analyzing the beautifulness of a photo. With the advancement of imaging technology, people capture and produce photos easily but they find it difficult to sort out good ones. Therefore, in this paper, we propose a method for efficient photo aesthetics analysis for quality classification. We developed some features based on human psychology and criterions on photography, including complexity, texture, color, composition. Particularly, the Sobel edge detector and Harris corner points are extracted to assess the image complexity, and some color and composition features proposed in previous literature are improved for better analysis accuracy.The supervised machine learning methods are applied for photo aesthetics analysis, and the SVM is used to measure the effectiveness of the proposed features. In our experiments, both image databases CUHKPQ and CUHK are used for evaluation. In the experiment, the CUHKPQ dataset is divided into seven thematic categories, whereas the CUHK dataset contains all categories. The experimental results show that the proposed method is able to classify the photo aesthetic quality efficiently with lower dimension compared to other related work. In the CHUK dataset, the accuracy above 88.72% and in the CUHKPQ dataset, the accuracy is up to 96.4%.

In this paper, a handwriting forgery detection system based on branchlet features and Gaussian mixture models(GMMs) is presented. At the beginning, the input handwriting images are processed by binarization and morphological operations to enhance sign traces. Subsequently, a thinning algorithm is employed to obtain the skeleton of character strokes, and find branchlet points in the skeleton image to extract handwriting features. Next, the feature data of input handwriting images are combined into some groups exhaustively to create their own GMMs. Then the similarity between each group and input handwriting images is measured. Through a voting system, the input handwriting images that possess higher similarity values are deemed as real signs. In a sequel, a new GMM is created by the real sign images to measure the similarity of all input handwriting images. Finally, we calculate the sample mean and standard deviation of the above measured similarity values. By incorporating these statistical terms, the input handwriting images whose similarity values are below an assigned threshold will be predicted as forgery. In the experiments, we adopt the IAM Handwriting Database that includes 657 writers' handwriting images whose resolution is fixed at 300dpi, and stored as 256 gray scale PNG images. The experimental results reveal that in case of 20% forgery our proposed system can reach up to 95% accuracy. When performing cross-validation under the input handwriting images consisting of 20% to 60% forgery, the system can reach at the average accuracy of 80.52%. This unsupervised learning scheme is effectively to detect the forged handwriting in scanned documents.

In this paper, we employ particle filters to dynamically locate a face and upper-limbs. To prevent from the disturbance caused by skin color regions, such as other naked parts of a human body, or some skin color-like objects in the background, we further take the motion cue as a feature during the tracking. Currently, we prescribe eight kinds of upper-limbs postures with reference to the characteristic of flag semaphore. The advantage is that we can utilize the relative positions of a face and two hands to recognize the postures easily. To achieve posture recognition, we evaluate three different classifiers using the machine learning methods: multi-layer perceptrons, support vector machines, and AdaBoost algorithms. The experimental results reveal that AdaBoost algorithms are the best one, which reach the accuracy rate of recognizing upper-limbs postures more than 95% and require much less training time than the other two do.

In this paper, a new neural architecture, the multisynapse neural network, is developed for constrained optimization problems, whose objective functions may include high-order, logarithmic, and sinusoidal forms, etc., unlike the traditional Hopfield networks which can only handle quadratic form optimization. Meanwhile, based on the application of this new architecture, a fuzzy bidirectional associative clustering network (FBACN), which is composed of two layers of recurrent networks, is proposed for fuzzy-partition clustering according to the objective-functional method. It is well known that fuzzy c-means is a milestone algorithm in the area of fuzzy c-partition clustering. All of the following objective-functional-based fuzzy c-partition algorithms incorporate the formulas of fuzzy c-means as the prime mover in their algorithms. However, when an application of fuzzy c-partition has sophisticated constraints, the necessity of analytical solutions in a single iteration step becomes a fatal issue of the existing algorithms. The largest advantage of FBACN is that it does not need analytical solutions. For the problems on which some prior information is known, we bring a combination of part crisp and part fuzzy clustering in the third optimization problem.

A new algorithm is presented for the piecewise linear approximation of two-dimensional digitized curves against a square grid. The algorithm utilizes an adaptive reduction procedure in two approximation phases to select the critical points of a digitized curve such that the deviation, from the digitized curve to its final approximated curve, is bounded by a uniform error tolerance. The time complexity of this algorithm is O(m/sup 2/) rather than O(n/sup 2/) on the theoretical plane. In the experiments of fixing the initial and the final processing points, the performance of the algorithm has been compared to those of three prominent other algorithms regarding the required number of critical points and the total execution time of the program. Of the four algorithms compared, the present algorithm consistently has the shortest execution time of the program, and it tends most to require as few critical points as the optimum algorithm that was tested.>