Yoshihiro Hagihara

Face detection from cluttered images using a polynomial neural network

Abstract Automatic detection of human faces from cluttered images is important for face recognition and security applications. This problem is challenging due to the multitude of variations and the confusion between face and background regions. This paper proposes a new face detection method using a polynomial neural network (PNN). To locate the human faces in an image, the local regions in multiscale sliding windows are classified by the PNN to two classes, namely, face and non-face. The PNN takes as inputs the binomials of the projection of the local image onto a feature subspace learned by principal component analysis (PCA). We investigated the influence of PCA on either the face samples or the pooled face and non-face samples. In addition, we integrate the distance from the feature subspace into the PNN to improve the detection performance. In experiments on images with complex backgrounds, the proposed method has produced promising results in terms of high detection rate and low false positive rate.

Optimal image feature set for detecting lung nodules on chest X-ray images

The performance of a computer-aided diagnosis system depends on the feature set used in it. This paper shows the results of image feature selection experiments. We evaluated 210 features to look for the optimum feature set. For the purpose, a forward stepwise selection approach was employed. The area under the receiver operating characteristic (ROC) curve was adopted to evaluate the performance of each feature set. Analysis of the optimally selected feature set is given and the experiments using 247 chest x-ray images are also shown.

Detection of rounded opacities on chest radiographs using convergence index filter

This paper presents a method to detect rounded opacities on digital chest radiographs. Based on a model of rounded opacities, three filters that evaluate the convergence degree of gradient vectors in the neighborhood of the pixel of interest are investigated to detect rounded opacities in the complex lung area. These filters do not depend on the contrast of the rounded opacity to its background. Their outputs depend on only the distribution of the gradient vector orientation. Experiments to test their performance have been performed. Their results show the effectiveness of the proposed filters. Especially, the iris filter and the adaptive ring filter are shown to be very effective in detecting cancerous tumor candidates. Another experiment to compare the performance of the iris filter and the Min-/spl Phi/DD filter which is one of the filters effective in detecting rounded opacities have been performed and the superiority of the proposed filters has been shown.

2-D direction histogram based entropic thresholding

Abstract Local image features are effective descriptors for image analysis and are also important cues for image segmentation. In this paper, we propose a novel entropic thresholding approach. This approach incorporates local features into a conventional entropic method to implement the thresholding. The local features are obtained from an orientation histogram to describe the edge property of the local neighborhood. To verify the performance of our method, thresholding was carried out on different types of images and compared with some well-known entropic approaches. Experimental results show that using the local edge property can give a better thresholding result.

Automatic Image Enhancement by Artificial Bee Colony Algorithm

With regard to the improvement of image quality, image enhancement is an important process to assist human with better perception. This paper presents an automatic image enhancement method based on Artificial Bee Colony (ABC) algorithm. In this method, ABC algorithm is applied to find the optimum parameters of a transformation function, which is used in the enhancement by utilizing the local and global information of the image. In order to solve the optimization problem by ABC algorithm, an objective criterion in terms of the entropy and edge information is introduced to measure the image quality to make the enhancement as an automatic process. Several images are utilized in experiments to make a comparison with other enhancement methods, which are genetic algorithm-based and particle swarm optimization algorithm-based image enhancement methods.

A fast morphological filtering algorithm

This paper proposes an algorithm for fast morphological filtering using structuring elements of arbitrary two-dimensional shape. The conventional fast morphological filtering has the constraint that the structuring element should be decomposable. The proposed method solves this problem by adding the result for the area which becomes a new object of calculation as a result of the motion of the structuring element and deleting the result for the area which goes outside the object for calculation. The proposed method has the feature that the calculation result is retained for each area deleted simultaneously from the object of calculation, and then morphological filtering is realized. Compared to the conventional fast algorithm, the constraint is reduced in the sense that undecomposable structuring elements can be used, and the method can be applied to multivalued images. Furthermore, the processing speed is independent of the density resolution of the considered image. The computation complexity is O(4rs) for the case of a circular structuring element, for example, where r is the radius of the structuring element and s is the number of pixels in the original image. This paper presents an experiment for comparison between the proposed algorithm and the conventional algorithm. The experiment reveals that the proposed algorithm is approximately twice as fast as the conventional fast method. An experiment is also performed for ring-shaped structuring elements, to which the conventional fast method is difficult to apply, and the result is compared to the conventional method without speed improvement. The results reveal that the processing time in the method without speed improvement increases when the number of pixels in the structuring element is increased, but it does not increase in the proposed method. Thus, the proposed method is more effective when the processing time required in the conventional method is long.

Development of electric rust preventive machining method system – safe water using for machining fluid: complete removal of bacteria (Legionella pneumophila) and assay

In manufacturing, machining fluid as cutting oil or grinding fluid is used for machining. This fluid contains several chemicals such as oil, surface active agent, extreme pressure agent (chlorine, phosphorus, sulphur, etc.), rust preventive agent, antiseptic agent, germicide and so on for improvement machining performance. However, it is not good for human body and environment. It is afraid that workers health hazard is occurred by fluid mist absorption in breath and splash contacting. In addition, waste fluid needs disposal treatment (incineration or coagulative precipitation and landfill, etc.) that is high cost and heavy environmental load. Therefore, environmental friendly and harmless machining is proposed for nanomanufacturing and green manufacturing. The electric rust preventive machining method system is developed in this investigation. This method system use only water (tap water, under ground water, industrial water) as machining fluid. Water only machining lead to greatly decreasing of waste fluid treatment and petroleum oil resources saving. In addition, complete removal of bacteria (Legionella pneumophila) for safe machining water by improvement water recycle system installed reverse osmosis membrane and ultraviolet radiation unit with non-using chemicals is evaluated by biological cultivation assay. And artificially contaminated tank water bacteria amount tendency is estimated.

Development of an Innovative Water Machining System Employing the Electric Rust Preventive Method - Precise Evaluation of Purity of the Refined Water with a Laser Turbidity Meter

Green manufacturing is important subject. In order to reduce waste disposal cost and environmental load, decreasing machining fluid that contains several chemicals such as oil, extreme pressure agent etc. is demanded. In this investigation, the electric rust preventive machining method system that uses only water as machining fluid have been developed. This paper mentioned about evaluation of recycled water quality. The refined water that is purified with developed water recycle system which is installed reverse osmosis membrane (RO) is too clean to evaluate by normal method till now. Using laser turbidity meter, precise water purification is evaluated precisely. Therefore, it is clarified that RO refined water turbidity (RO1=0.0006, RO2=0.0003) is very low compared with clean tap water (0.1207). So, water recycle system can remove contaminating fine particle from water. It is expected that scratch less ultra precision machining with water is enabled to conduct with high filtration ability.

Development of Electric Rust Preventive Machining Method - Improvement of Electric Rust Preventive Chip Sedimentation System

The machining (cutting, grinding etc.) is conducted in manufacturing. Machining fluid (cutting oil, grinding fluid) that consists of oil, surface active agent, and extreme pressure agent, anti rust agent etc. is used. It improves machining performance, but it needs waste fluid disposal that is incineration or coagulative precipitation and so on. It causes huge cost and environmental load. Furthermore, it is afraid of workers health hazard for several chemicals while machining. Therefore, the electric rust preventive machining method system (water machining) is proposed and developed. This method uses only harmless water (tap water etc.) as machining fluid. In this paper, improvement of electric rust preventive chip sedimentation system that is part of water recycle system which is used for machining water purification and re-use. On long time (3 days) preservation of iron chip in water, decreasing of rust and turbidity is examined. Improved electric rust preventive chip sedimentation system is equipped simple circulation filter unit newly and its effectiveness is clarified. When many quantity of iron powder (3kg) likened to actual sludge is sunken in sedimentation water tank, if electric rust prevention and simple circulation filter activated, turbidity and colour would be decreased greatly. Therefore, it is expected that purification load of next part of filters is decreased and life-time of filter and system will be prolonged.

Development of Electric Rust Preventive Machining Method – Optimization of Water Recycle System

In production site, machining fluid (cutting oil, grinding fluid) is used. It contains several chemicals that are oil, surface active agent, and extreme pressure agent, anti rust agent and so on. Waste fluid disposal which is incineration etc. is necessary and arise huge cost and environmental load. In addition, workers health hazard is concerned for several chemicals while machining. In this investigation, new water machining method system (electric rust preventive machining method system) that uses only water as machining fluid for solving of conventional machining fluid problems is developed. In particularly, this paper mentions optimization of used machining water recycle on purification rate and refined water flow quantity in developed water recycle system. Therefore, high speed adjustment test liquid equipment is developed for stable experimental condition for evaluation. Test liquid turbidity is random for sludge particle and simple filter decreases this fluctuation. However, water recycle system is aimed for constant refined output despite fluctuation of input dirty water, and it is achieved. The optimized refined water flow quantity is 13.3 L/min at 1.0 MPa from viewpoint of purification on iron, turbidity, colour, conductivity and flow rate and purification load for reverse osmosis membrane.