Confluence is a novel non-Intersection over Union (IoU) alternative to Non-Maxima Suppression (NMS) in bounding box post-processing in object detection. It overcomes the inherent limitations of IoU-based NMS variants to provide a more stable, consistent predictor of bounding box clustering by using a normalized Manhattan Distance inspired proximity metric to represent bounding box clustering. Unlike Greedy and Soft NMS, it does not rely solely on classification confidence scores to select optimal bounding boxes, instead selecting the box which is closest to every other box within a given cluster and removing highly confluent neighboring boxes. Confluence is experimentally validated on the MS COCO and CrowdHuman benchmarks, improving Average Precision by 0.2--2.7% and 1--3.8% respectively and Average Recall by 1.3--9.3 and 2.4--7.3% when compared against Greedy and Soft-NMS variants. Quantitative results are supported by extensive qualitative analysis and threshold sensitivity analysis experiments support the conclusion that Confluence is more robust than NMS variants. Confluence represents a paradigm shift in bounding box processing, with potential to replace IoU in bounding box regression processes.

In order to separate the chromatogram peaks and spectra from the High Performance Liquid Chromatography with Diode Array Detector (HPLC-DAD) data set, a separation model of Generalized Reference Curve Measurement and its solution by multitarget Bare Bones Particle Swarm Optimization (GRCMmBBPSO) is proposed in this paper. Firstly, parameters are constructed which will generate Reference Curves (RCs) for chromatogram peaks. Secondly, the GRCM model is proposed to transform all these parameters to scalar values, which indicate the possibility for the HPLC-DAD data set containing chromatogram peaks similar to the RCs constructed by corresponding parameters. Then, the algorithm of mBBPSO is introduced to calculate the optimal parameters by minimizing the scalar values. Finally, the spectra are estimated. Through simulations and experiments, following conclusions are drawn: (1) the GRCMmBBPSO method can extract chromatogram peaks from the simulation data set without knowing the number of the compounds in advance even when a severe overlap and white noise exist; (2) the GRCMmBBPSO method can be applied to real HPLC-DAD data set.

Query spotting in document images is a subclass of Content-Based Image Retrieval (CBIR) algorithms concerned with detecting occurrences of a query in a document image. Due to noise and complexity of document images, spotting can be a challenging task and easily prone to false positives and partially incorrect matches, thereby reducing the overall precision of the algorithm. A robust and accurate spotting algorithm is essential to our current research on sketch-based retrieval of digitized lecture materials. We have recently proposed a modular spotting algorithm in [1]. Compared to existing methods, our algorithm is both application-independent and segmentation-free. However, it faces the same challenges of noise and complexity of images. In this paper, inspired by our earlier research on optimizing parameter settings for CBIR using an evolutionary algorithm [2][3], we introduce a Genetic Algorithm-based optimization step in our spotting algorithm to improve each spotting result. Experiments using an image dataset of journal pages reveal promising performance, in that the precision is significantly improved but without compromising the recall of the overall spotting result.

We report our ongoing research on an application-independent and segmentation-free approach for spotting queries in document images. Built on our earlier work reported in [1][2], this paper introduces an image processing approach that finds occurrences of a query, which is a multi-part object, in a document image, through 5 steps: (1) Preprocessing for image normalization and connected components extraction. (2) Feature Extraction from connected components. (3) Matching of the query and document image connected components' feature vectors. (4) Voting for determining candidate occurrences in the document image that are similar to the query. (5) Candidate Filtering for detecting relevant occurrences and filtering out irrelevant patterns. Compared to existing methods, our contributions are twofold: Our approach is designed to deal with any type of queries, without restriction to a particular class such as words or mathematical expressions. Second, it does not apply a domain-specific segmentation to extract regions of interest from the document image, such as text paragraphs or mathematical calculations. Instead, it considers all the image information. Experimental evaluation using scanned journal images show promising performances and possibility of further improvement.

BACKGROUND: The 3D chromatogram generated by High Performance Liquid Chromatography-Diode Array Detector (HPLC-DAD) has been researched widely in the field of herbal medicine, grape wine, agriculture, petroleum and so on. Currently, most of the methods used for separating a 3D chromatogram need to know the compounds' number in advance, which could be impossible especially when the compounds are complex or white noise exist. New method which extracts compounds from 3D chromatogram directly is needed. METHODS: In this paper, a new separation model named parallel Independent Component Analysis constrained by Reference Curve (pICARC) was proposed to transform the separation problem to a multi-parameter optimization issue. It was not necessary to know the number of compounds in the optimization. In order to find all the solutions, an algorithm named multi-areas Genetic Algorithm (mGA) was proposed, where multiple areas of candidate solutions were constructed according to the fitness and distances among the chromosomes. RESULTS: Simulations and experiments on a real life HPLC-DAD data set were used to demonstrate our method and its effectiveness. Through simulations, it can be seen that our method can separate 3D chromatogram to chromatogram peaks and spectra successfully even when they severely overlapped. It is also shown by the experiments that our method is effective to solve real HPLC-DAD data set. CONCLUSIONS: Our method can separate 3D chromatogram successfully without knowing the compounds' number in advance, which is fast and effective.

In order to separate a 3D chromatography, which is generated from High Performance Liquid Chromatography-Diode Array Detector (HPLC-DAD), into chromatograms and spectra, we proposed a model called parallel Independent Component Analysis constrained by Reference Curve (pICARC), which transforms the separation problem to a multi-parameter optimization issue. Then, A new algorithm named multi-areas Genetic Algorithm (mGA) is developed to search multiple solutions in parallel. It was demonstrated that our approach could successfully separate a given HPLC-DAD dataset into chromatograms and spectra with little errors and in short computational time.

Content-based Image Retrieval (CBIR) has been an active area of research for retrieving similar images from large repositories, without the prerequisite of manual labeling. Most current CBIR algorithms can faithfully return a list of images that matches the visual perspective of their inventors, who might decide to use a certain combination of image features like edges, colors and textures of regions as well as their spatial distribution during processing. In practice, however, the retrieved images rarely correspond exactly to the results expected by the users, a problem that has come to be known as the semantic gap. In this paper, we propose a novel and extensible multidimensional approach called matrix of visual perspectives as a solution for addressing this semantic gap. Our approach exploits the dynamic cross-interaction (in other words, mix-and-match) of image features and similarity metrics to produce results that attempt to mimic the mental visual picture of the user. Experimental results on retrieving similar Japanese cultural heritage symbols called kamons by a prototype system confirm that the interaction of visual perspectives in the user can be effectively captured and reflected. The benefits of this approach are broader. They can be equally applicable to the development of CBIR systems for other types of images, whether cultural or noncultural, by adapting to different sets of application specific image features.

In recent years, law enforcement personnel have greatly been aided by the deployment of Automated Fingerprint Identification Systems (AFIS). These systems largely operate by matching salient features automatically extracted from fingerprint images for their decision. However, there are two major shortcomings in current systems. First, the result of identification depends primarily on the chosen features and the algorithm that matches them. Second, these systems cannot improve their results by benefiting from interactions with seasoned examiners who often can identify minute differences between fingerprints beyond that is capable of by current systems. In this paper, we propose a system for fingerprint identification that incorporates relevance feedback. We show that a persistent semantic space over the database of fingerprints can be incrementally learned. Here, the learning module makes use of a dimensionality reduction process that returns both a low-dimensional semantic space and an out-of-sample mapping function, achieving a two-fold benefits of data compression and the ability to project novel fingerprints directly onto the semantic space for identification. Experimental results demonstrated the potential of this learning framework for adaptive fingerprint identification.

Using local data information, the recently proposed local Fisher Discriminant Analysis (LFDA) algorithm18 provides a new way of handling the multimodal issues within classes where the conventional Fisher Discriminant Analysis (FDA) algorithm fails. Like the FDA algorithm (global counterpart), the LFDA suffers when it is applied to the higher dimensional data sets. In this paper, we propose a new formulation by which a robust algorithm can be formed. The new algorithm offers more robust results for higher dimensional data sets when compared with the LFDA in most cases. By extensive simulation studies, we have demonstrated the practical usefulness and robustness of our new algorithm in data visualization.

In most existing dimensionality reduction algorithms, the main objective is to preserve relational structure among objects of the input space in a low dimensional embedding space. This is achieved by minimizing the inconsistency between two similarity/dissimilarity measures, one for the input data and the other for the embedded data, via a separate matching objective function. Based on this idea, a new dimensionality reduction method called Twin Kernel Embedding (TKE) is proposed. TKE addresses the problem of visualizing non-vectorial data that is difficult for conventional methods in practice due to the lack of efficient vectorial representation. TKE solves this problem by minimizing the inconsistency between the similarity measures captured respectively by their kernel Gram matrices in the two spaces. In the implementation, by optimizing a nonlinear objective function using the gradient descent algorithm, a local minimum can be reached. The results obtained include both the optimal similarity preserving embedding and the appropriate values for the hyperparameters of the kernel. Experimental evaluation on real non-vectorial datasets confirmed the effectiveness of TKE. TKE can be applied to other types of data beyond those mentioned in this paper whenever suitable measures of similarity/dissimilarity can be defined on the input data.

The paper proposes a new publishing system using PC and large-sized printer that can produce high quality printed matter comprising text, charts, illustrations, etc. As the core component of this system, a coding method that can ensure the quality of decoded images even on affine-transformed enlargement or reduction is needed. Conventional coding methods, however, cannot claim to have satisfied such a requirement. The paper presents a novel method for encoding and decoding binary images via adaptive approximation of the object contours using fluency functions, in the form of straight lines, arcs, and curved lines, that were proposed by one of the authors and others (see Kamada, M. et al., J. Approx. Theory, vol.55, no.1, p.27-34, 1988). Based on experimental results, we demonstrate that the quality of the decoded images is maintained without jaggy-noise, even on resizing. Furthermore, signboards produced by this publishing system have been used in several live events.

In this paper, a Trademark Application Assistant (TAST) that aims to speed up the process of a successful trademark application is proposed. The core of TAST is a similarity-based image retrieval system that makes use of relaxation matching on extracted contour segments. A novelty of this research lies in the processing architecture introduced for adapting the highly effective but computation expensive relaxation matching algorithm to real-time image retrieval. Preliminary results illustrating the quality of matching and the scalability of the processing architecture in terms of response time are included. The proposed architecture is not limited to using relaxation matching based servers, but is also applicable to other types of matching server as well.

The authors' previous work in image similarity retrieval by relaxation labeling processes (2001) required adding local constraints to obtain an initial set of compatible objects and labels on pairs of images to ensure labeling consistency at convergence. This approach suffers from the problem of over-specified constraints that leads to potentially invaluable information being prematurely removed. To address this problem, we introduce the idea of a Match-all label for objects that failed these constraints. It serves to give them a defined labeling probability as well as allowing them participate in global correspondences through the compatibility model. We show that this enhanced formulation still meets the conditions for a theorem on labeling consistency in Hummel and Zucker (1983) to be satisfied. At convergence, the set of objects and their most consistent labels constitute a best partial labeling for the pair of images.