It is very challenging to accurately detect smoke from images because of large variances of smoke colour, textures, shapes and occlusions. To improve performance, the authors combine dual threshold AdaBoost with staircase searching technique to propose and implement an image smoke detection method. First, extended Haar-like features and statistical features are efficiently extracted from integral images from both intensity and saturation components of RGB images. Then, a dual threshold AdaBoost algorithm with a staircase searching technique is proposed to classify the features of smoke for smoke detection. The staircase searching technique aims at keeping consistency of training and classifying as far as possible. Finally, dynamic analysis is proposed to further validate the existence of smoke. Experimental results demonstrate that the proposed system has a good robustness in terms of early smoke detection and low false alarm rate, and it can detect smoke from videos with size of 320 × 240 in real time.

As one of the key technologies in ultrasound (US)-guided high-intensity focused ultrasound (HIFU) ablation systems, a precise and automatic US image segmentation method for tumour localisation can contribute to ablation of the diseased tissues and avoiding unwanted destruction of the healthy tissues. Owing to the speckle noise and the irregular shapes of target tumours in US images, traditional image segmentation methods are not suitable for tumour localisation. In this study, the authors proposed an improved gradient and direction vector flow (G&DVF) model to segment US images for tumour localisation. The conventional G&DVF model was improved in three aspects to obtain a better segmentation for tumour localisation. The straight lines were changed into fold lines to increase their flexibility in guiding the active contour. Several weighting parameters were added to the energy functional to control the influences of these different lines. Moreover, a novel vector field was defined to reduce the computational complexity and save the computation time. With these three aspects improved, the improved G&DVF model was applied in US image segmentation for tumour localisation. The experimental results demonstrate that this proposed model is reliable, accurate and time saving in US image segmentation for tumour localisation.

The existing non-hybrid image inpainting techniques can be broadly classified into two types. One is the texture-based inpainting and the other is the structure-based inpainting. One critical drawback of those techniques is that their inpainting results are not effective for the images with a mixture of texture and structure features in terms of visual quality or processing time. However, the conventional hybrid inpainting algorithms, which aim at inpainting images with texture and structure features, do not effectively deal with the two items: (i) what is the most effective application order of the constituents? and (ii) how can one extract a minimal sub-image that may contain best candidates of inpainting source? In this study, the authors propose a new hybrid inpainting algorithm to address the two tasks fully and effectively. Precisely, the authors’ algorithm attempts to solve two key ingredients: (i) (right time) determining the best application order for inpainting textural and structural missing regions and (ii) (right place) extracting the sub-image containing best candidates of source patches to be used to fill in a target region.

Handwritten digit recognition is one of the challenging problems of character recognition because of the large variation in writing styles of individuals and the presence of similar looking shapes of different numerals. Most of the feature extraction techniques are based on statistical or topological attributes of the image in its spatial domain, barring few works attempting feature extraction in a transformed domain. Another challenge is the optimal selection of zones while extracting features from localised zones of the unknown (test) image. In most of the cases, the recognition phase, being isolated from the training phase makes it impossible to adaptively improve the feature selection using the knowledge obtained from error analysis. In this study, the authors propose a feature extraction technique, new to the character recognition problem, using non-redundant Stockwell transform. Another transformed domain feature extraction using Slantlet coefficients is proposed. They also propose to use bio-inspired and evolutionary computing-based optimisation techniques to adaptively select the optimal zone arrangement in the feature selection stage from the knowledge of classification accuracy. The proposed methods are experimentally validated on handwritten digit database of Odia language which proves to outperform any recognition accuracy reported before.

Structure fusion (SF) has been presented for multiple feature fusion via mining the discriminative and complementary information from different feature sets. As the typical methods, SF based on locality preserving projections (SFLPP) and SF based on tensor subspace analysis (SFTSA) have been developed for classification by capturing the complete structure from different features. However, the jointed optimisation function of SFLPP or SFTSA does not clearly explain the modelling mechanism of SF, and its solving process is complex because of iterative eigenvalue decomposition. In this study, structure modelling based on maximisation posterior probability (SMMPP) is proposed for solving these issues. It jointly considers both the certain prior structure (the mutual structure of multiple feature structure described by Ising model) and the uncertain likelihood structure (the possible fusion structure of multiple feature structure represented by Markov random field model) into the framework of Bayes’ rule. The proposed computational solution is faster-converging speed than SFLPP or SFTSA with the guarantee of convergence. Extensive experiments conducted on shape analysis and human action recognition demonstrate the superiority of SMMPP over the state of art methods.

A fundamental problem of image registration is the determination of corresponding points. The scale-invariant feature transform (SIFT) is a well-known algorithm in this regard. However, SIFT suffers from quantity, quality and distribution of the detected points when facing with high-resolution and low-contrast colour retinal fundus images. This study introduces an improved SIFT algorithm which identifies adequate, stable and distinctive keypoints with uniform distribution in the overlapped areas. The keypoint of the proposed method is a selection strategy of the difference of Gaussian (DoG) extremum points according to a stability score to guarantee the feature qualities. The stability score is based on the DoG values of extremum points and their vesselness measures in the relevant Gaussian images. Since the selected points lie on the vessels, which are relatively stable between image pairs, the points are unaffected by illumination and content variations of retinal backgrounds. The detected points are introduced to an integrated outlier rejection method. Then, the correspondences determine the geometric transformation parameters. The authors examined quantitatively and qualitatively the performance of this algorithm on four datasets including temporal and partially overlapping image pairs. The experimental results show the outperformance of the approach over similar methods in terms of efficiency, positional accuracy and speed.

An improved centroid extraction algorithm for autonomous star sensor is proposed in this study, which focuses on the improvements of the location accuracy of stars and the speed of the centroid extraction. First, the coarse positioning of stars is carried out to achieve the dispersive regions of the stars quickly. Then the stars pixels are chosen by the automatic seeded region growing algorithm. Subsequently, in order to restrain the interference of noise, the grey values of the stars pixels are modified according to the characteristics of the star energy distribution. Finally, the fine positioning of the star can be achieved using the proposed centroid calculation formula. Experimental results show that the proposed algorithm has high-positioning accuracy and good noise resistant ability compared with the other two centroid extraction algorithms. Moreover, the computational complexity of the proposed algorithm is lower than that of the other two algorithms.

Image enhancement has an important role in image processing applications. Contrast limited adaptive histogram equalisation (CLAHE) is an effective algorithm to enhance the local details of an image. However, it faces the contrast overstretching and noise enhancement problems. To solve these problems, this study presents a novel image enhancement method, named CLAHE-discrete wavelet transform (DWT), which combines the CLAHE with DWT. The new method includes three main steps: First, the original image is decomposed into low-frequency and high-frequency components by DWT. Then, the authors enhance the low-frequency coefficients using CLAHE and keep the high-frequency coefficients unchanged to limit noise enhancement. This is because the high-frequency component corresponds to the detail information and contains most noises of original image. Finally, reconstruct the image by taking inverse DWT of the new coefficients. To alleviate over-enhancement, the reconstructed and original images are averaged using an originally proposed weighting factor. The weighting operation can control the enhancement levels of regions with different luminances in original image adaptively. This is important because bright parts of image are usually needless to be enhanced in comparison with the dark parts. Extensive experiments show that this method performs well in detail preservation and noise suppression.

Spectral matching (SM) is an efficient and effective greedy algorithm for solving the graph matching problem in feature correspondence in computer vision and graphics. However, the classic SM algorithm cannot extract correspondences well when the affinity matrix is sparse and reducible (i.e. its corresponding graph is not connected). This case often happens when the geometric deformations consist of transformations with local inconsistency. The authors analyse this problem and show how the original SM could fail in this scenario. Then, the authors propose a revised two-step pipeline to tackle this issue: (1) decompose the mutually inconsistent local deformations into several consistent transformations which can be solved by individual SM; (2) filter out incorrect correspondences through an automatic thresholding. The authors perform experiments to demonstrate that this modification can effectively handle the coarse correspondence computation in shape or image registration where the global transformation consists of multiple inconsistent local transformations.