Object Image Classification Research Articles

An Object Classification Approach Based on Randomized Visual Vocabulary and Clustering Aggregation

Considering the problems with the conventional Bag-of-Visual-Words approaches, such as high time consumption, the synonymy and ambiguity of visual word, and instability of clustering high-dimensionality image local features, this paper presents a novel object classificaiton approach based on randomized visual vocabulary and clustering aggregation. Firstly, Exact Euclidean Locality Sensitive Hashing (E2LSH) is used to cluster local features of the training dataset, and a group of randomized visual vocabularies is constructed. Then, the randomized visual vocabularies are aggregated using clustering aggregation technique, resulting in Randomized Visual Vocabularies Aggregating Dictionary (RVVAD). Finally, the visual words histogram is generated according to the dictionary, and the Support Vector Machines are learned to accomplish image object categorization. Experimental results indicate that the expression ability of the dictionary is effectively improved, and the object classification precision is increased dramatically.

Object-based land cover change detection for cross-sensor images

Accurate and timely land cover change detection at regional and global scales is necessary for both natural resource management and global environmental change studies. Satellite remote sensing has been widely used in land cover change detection over the past three decades. The variety of satellites which have been launched for Earth Observation (EO) and the large volume of remotely sensed data archives acquired by different sensors provide a unique opportunity for land cover change detection. This article introduces an object-based land cover change detection approach for cross-sensor images. First, two images acquired by different sensors were stacked together and principal component analysis (PCA) was applied to the stacked data. Second, based on the Eigen values of the PCA transformation, six principal bands were selected for further image segmentation. Finally, a land cover change detection classification scheme was designed based on the land cover change patterns in the study area. An image–object classification was implemented to generate a land cover change map. The experiment was carried out using images acquired by Landsat 5 TM and IRS-P6 LISS3 over Daqing, China. The overall accuracy and kappa coefficient of the change map were 83.42% and 0.82, respectively. The results indicate that this is a promising approach to produce land cover change maps using cross-sensor images.

Histogram Curve Matching Approaches for Object-based Image Classification of Land Cover and Land Use.

The classification of image-objects is usually done using parametric statistical measures of central tendency and/or dispersion (e.g., mean or standard deviation). The objectives of this study were to analyze digital number histograms of image objects and evaluate classifications measures exploiting characteristic signatures of such histograms. Two histograms matching classifiers were evaluated and compared to the standard nearest neighbor to mean classifier. An ADS40 airborne multispectral image of San Diego, California was used for assessing the utility of curve matching classifiers in a geographic object-based image analysis (GEOBIA) approach. The classifications were performed with data sets having 0.5 m, 2.5 m, and 5 m spatial resolutions. Results show that histograms are reliable features for characterizing classes. Also, both histogram matching classifiers consistently performed better than the one based on the standard nearest neighbor to mean rule. The highest classification accuracies were produced with images having 2.5 m spatial resolution.

Assessing the spatio-temporal rates and patterns of land-use and land-cover changes in the Cerrados of southeastern Mato Grosso, Brazil

The Cerrados of central Brazil have undergone profound landscape transformation in recent decades due to agricultural expansion, and this remains poorly assessed. The present research investigates the spatial-temporal rates and patterns of land-use and land-cover (LULC) changes in one of the main areas of agricultural production in Mato Grosso State (Brazil), the region of Primavera do Leste. To quantify the different aspects of LULC changes (e.g. rates, types, and spatial patterns) in this region, we applied a post-classification change detection method, complemented with landscape metrics, for three dates (1985, 1995, and 2005). LULC maps were obtained from an object-based classification approach, using the nearest neighbour (NN) classifier and a multi-source data set for image object classification (e.g. seasonal Thematic Mapper (TM) bands, digital elevation model (DEM), and a Moderate Resolution Imaging Spectroradiometer (MODIS)-derived index), strategically chosen to increase class separability. The results provided an improved mapping of the Cerrados natural vegetation conversion into crops and pasture once auxiliary data were incorporated into the classification data set. Moreover, image segmentation was crucial for LULC map quality, in particular because of crop size and shape. The changes detected point towards increasing loss and fragmentation of natural vegetation and high rates of crop expansion. Between 1985 and 2005, approximately 42% (6491 km2) of Cerrados in the study area were converted to agricultural land uses. In addition, it was verified that cultivated areas are encroaching into fragile environments such as wetlands, which indicates the intense pressure of agricultural expansion on the environment.

Integration of orthoimagery and lidar data for object-based urban thematic mapping using random forests

Using high-spatial-resolution multispectral imagery alone is insufficient for achieving highly accurate and reliable thematic mapping of urban areas. Integration of lidar-derived elevation information into image classification can considerably improve classification results. Additionally, traditional pixel-based classifiers have some limitations in regard to certain landscape and data types. In this study, we take advantage of current advances in object-based image analysis and machine learning algorithms to reduce manual image interpretation and automate feature selection in a classification process. A sequence of image segmentation, feature selection, and object classification is developed and tested by the data sets in two study areas (Mannheim, Germany and Niagara Falls, Canada). First, to improve the quality of segmentation, a range image of lidar data is incorporated in an image segmentation process. Among features derived from lidar data and aerial imagery, the random forest, a robust ensemble classifier, is then used to identify the best features using iterative feature elimination. On the condition that the number of samples is at least two or three times the number of features, a segmentation scale factor has no particular effect on the selected features or classification accuracies. The results of the two study areas demonstrate that the presented object-based classification method, compared with the pixel-based classification, improves by 0.02 and 0.05 in kappa statistics, and by 3.9% and 4.5% in overall accuracy, respectively.

Automatic target classification of man-made objects in synthetic aperture radar images using Gabor wavelet and neural network

Processing of synthetic aperture radar (SAR) images has led to the development of automatic target classification approaches. These approaches help to classify individual and mass military ground vehicles. This work aims to develop an automatic target classification technique to classify military targets like truck/tank/armored car/cannon/bulldozer. The proposed method consists of three stages via preprocessing, feature extraction, and neural network (NN). The first stage removes speckle noise in a SAR image by the identified frost filter and enhances the image by histogram equalization. The second stage uses a Gabor wavelet to extract the image features. The third stage classifies the target by an NN classifier using image features. The proposed work performs better than its counterparts, like K-nearest neighbor (KNN). The proposed work performs better on databases like moving and stationary target acquisition and recognition against the earlier methods by KNN.

Corner Detection and Classification of Simple Objects in Low-Depth Resolution Range Images

This paper deals with corner detection of simple geometric objects in quantized range images. Low depth resolution and noise introduce challenges in edge and corner detection. Corner detection and classification is based on layer by layer depth data extraction and morphologic operations. Appearance based heuristics are applied to identify different corner types defined in this paper. Both computer generated and captured range images are dealt with. Synthetic range images have arbitrary range resolution while captured images are based on the sensor used. Real world data is collected using a structured light based sensor to provide dense range map.

Chinese Character Recognition by Tchebichef Moment Features

Abstract — Moment descriptors have long been applied in object recognition since the early years of the development of the moment theories. Nowadays, discrete orthogonal moments have been studied and proposed for they are superior to traditional continuous ones. In this paper, a set of moment features extracted from the discrete Tchebichef moments for Chinese character recognition is presented. A new method of evaluating the variance values of each moment feature is applied in this research. Tested on a set of 6,763 Chinese characters, our newly proposed Tchebichef moment features perform very well in distinguishing all Chinese character pairs that have similar structures. Index Terms — Discrete orthogonal moments, tchebichef moments, Chinese character recognition. I. I NTRODUCTION Since Hu [1] introduced the moment methods in 1961, moment descriptors have been widely used in image representation, pattern recognition, and object classification. One of the applications using moment features is the Chinese character recognition. Chinese characters are very different from many other languages in conveying information, while the structure of a character is a key to its meaning and pronunciation. Many feature extraction methods applied in the Chinese character recognition systems are based on the local features, such as strokes and feature points [2]-[5]. While the existing methods are quite efficient in general, there are some difficulties to distinguish two characters when they have very close structures. On the other hand, the moment method has the advantage of utilizing the global features of the Chinese characters, therefore, it can be used as a complementary scheme to overcome the obstacles confronted by other systems. Some Chinese character recognition systems based on the orthogonal moment descriptors have been reported [6]-[8]. Recently, the emergence of discrete orthogonal moments has substantially enriched the moment methods. Moments based on Tchebichef and Krawtchouk polynomials were introduced by Mukundan

Brain Tumor Classification Using Neural Network Based Methods

MRI (Magnetic resonance Imaging) brain tumor images Classification is a difficult task due to the variance and complexity of tumors. This paper presents two Neural Network techniques for the classification of the magnetic resonance human brain images. The proposed Neural Network technique consists of three stages, namely, feature extraction, dimensionality reduction, and classification. In the first stage, we have obtained the features related with MRI images using discrete wavelet transformation (DWT). In the second stage, the features of magnetic resonance images (MRI) have been reduced using principles component analysis (PCA) to the more essential features. In the classification stage, two classifiers based on supervised machine learning have been developed. The first classifier based on feed forward artificial neural network (FF-ANN) and the second classifier based on Back-Propagation Neural Network. The classifiers have been used to classify subjects as normal or abnormal MRI brain images. Artificial Neural Networks (ANNs) have been developed for a wide range of applications such as function approximation, feature extraction, optimization, and classification. In particular, they have been developed for image enhancement, segmentation, registration, feature extraction, and object recognition and classification. Among these, object recognition and image classification is more important as it is a critical step for high-level processing such as brain tumor classification. Multi-Layer Perceptron (MLP), Radial Basis Function (RBF), Hopfield, Cellular, and Pulse-Coupled neural networks have been used for image segmentation. These networks can be categorized into feed-forward (associative) and feedback (auto-associative) networks..

Deformable Bayesian Network: A Robust Framework for Underwater Sensor Fusion

The dynamic tree (DT) graphical model is a popular analytical tool for image segmentation and object classification tasks. A DT is a useful model in this context because its hierarchical property enables the user to examine information in multiple scales and its flexible structure can more easily fit complex region boundaries compared to rigid quadtree structures such as tree-structured Bayesian networks. This paper proposes a novel framework for data fusion called a deformable Bayesian network (DFBN) by using a DT model to fuse measurements from multiple sensing platforms into a nonredundant representation. The structural flexibility of the DFBN will be used to fuse common information across different sensor measurements. The appropriate structure update strategies for the DFBN and its parameters for the data fusion application are discussed. A real-world example application using sonar images collected from a survey mission is presented. The fusion results using the presented DFBN framework are shown to outperform state-of-the-art approaches such as the Gaussian mean shift and spectral clustering algorithms. The DFBN's complexity and scalability are discussed to address its potential for a larger data set.

A SURVEY OF CORE TECHNOLOGIES TO BUILD AN INTELLIGENT SURVEILLANCE SYSTEM

Public Security is a great concern past few years in INDIA as well as the World due to terrorist activities. It has been found that to enhance security a network of cameras is being deployed across the prominent places in the world. Such a traditional approach is costly as well as totally depends on physical and mental status of the observer looking at the camera data. In the light of such weak security infrastructure we propose a method to develop an intelligent system which not only will capture and feedback the video at real time but also model human behavior. To develop an Intelligent System for detecting , Modeling , Classification of human behavior using image processing and machine vision to detect abnormal malicious intents. We have identified the necessary steps and modules to achieve our proposed system. In the light of the study we have anticipated various algorithms and provide a survey of work done in the fields of image processing and video processing. We have identified important and core technologies as Image segmentation, Background removal, Object classification, motion tracking and behavior analysis and have presented each technology with its recent developments. Index Terms : modeling , image processing, machine vision, image segmentation, background removal, tracking.

Object-Based Image Analysis of Downed Logs in Disturbed Forested Landscapes Using Lidar

Downed logs on the forest floor provide habitat for species, fuel for forest fires, and function as a key component of forest nutrient cycling and carbon storage. Ground-based field surveying is a conventional method for mapping and characterizing downed logs but is limited. In addition, optical remote sensing methods have not been able to map these ground targets due to the lack of optical sensor penetrability into the forest canopy and limited sensor spectral and spatial resolutions. Lidar (light detection and ranging) sensors have become a more viable and common data source in forest science for detailed mapping of forest structure. This study evaluates the utility of discrete, multiple return airborne lidar-derived data for image object segmentation and classification of downed logs in a disturbed forested landscape and the efficiency of rule-based object-based image analysis (OBIA) and classification algorithms. Downed log objects were successfully delineated and classified from lidar derived metrics using an OBIA framework. 73% of digitized downed logs were completely or partially classified correctly. Over classification occurred in areas with large numbers of logs clustered in close proximity to one another and in areas with vegetation and tree canopy. The OBIA methods were found to be effective but inefficient in terms of automation and analyst’s time in the delineation and classification of downed logs in the lidar data.

3D Model-Based Semantic Categorization of Still Image 2D Objects

Automatic classification and interpretation of objects present in 2D images is a key issue for various computer vision applications. In particular, when considering image/video, indexing, and retrieval applications, automatically labeling in a semantically pertinent manner/huge multimedia databases still remains a challenge. This paper examines the issue of still image object categorization. The objective is to associate semantic labels to the 2D objects present in natural images. The principle of the proposed approach consists of exploiting categorized 3D model repositories to identify unknown 2D objects, based on 2D/3D matching techniques. The authors use 2D/3D shape indexing methods, where 3D models are described through a set of 2D views. Experimental results, carried out on both MPEG-7 and Princeton 3D models databases, show recognition rates of up to 89.2%.

Object-oriented mapping of landslides using Random Forests

Landslide inventory mapping is an indispensable prerequisite for reliable hazard and risk analysis, and with the increasing availability of very high resolution (VHR) remote sensing imagery the creation and updating of such inventories on regular bases and directly after major events is becoming possible. The diversity of landslide processes and spectral similarities of affected areas with other landscape elements pose major challenges for automated image processing, and time-consuming manual image interpretation and field surveys are still the most commonly applied mapping techniques. Taking advantage of recent advances in object-oriented image analysis (OOA) and machine learning algorithms, a supervised workflow is proposed in this study to reduce manual labor and objectify the choice of significant object features and classification thresholds. A sequence of image segmentation, feature selection, object classification and error balancing was developed and tested on a variety of sample datasets (Quickbird, IKONOS, Geoeye-1, aerial photographs) of four sites in the northern hemisphere recently affected by landslides (Haiti, Italy, China, France). Besides object metrics, such as band ratios and slope, newly introduced topographically-guided texture measures were found to enhance significantly the classification, and also feature selection revealed positive influence on the overall performance. With an iterative procedure to examine the class-imbalance within the training sample it was furthermore possible to compensate spurious effects of class-imbalance and class-overlap on the balance of the error rates. Employing approximately 20% of the data for training, the proposed workflow resulted in accuracies between 73% and 87% for the affected areas, and approximately balanced commission and omission errors.

Dense statistic versus sparse feature-based approach for 3D object recognition

In this article we introduce and compare two approaches towards automatic classification of 3D objects in 2D images. The first one is based on statistical modeling of wavelet features. It estimates probability density functions for all possible object classes considered in a particular recognition task. The second one uses sparse local features. For training, SURF features are extracted from the training images. During the recognition phase, features from the image are matched geometrically, providing the best fitting object for the query image. Experiments were performed for different training sets using more than 40 000 images with different backgrounds. Results show very good classification rates for both systems and point out special characteristics for each approach, which make them more suitable for different applications.

Fast recognition of foreign fibers in cotton lint using machine vision

This paper presents an approach for fast segmentation of foreign fiber images and precise recognition of foreign fiber objects using machine vision. Live images were acquired in real time using a line scan CCD camera. After an image was acquired it was transferred to a host computer immediately for image processing and object classification. The captured image was firstly segmented according to the mean and standard deviation of R, G and B values of each pixel in the image. Then noises were removed using the area threshold method. Afterwards, color features, shape features and texture features of each foreign fiber object were extracted. Finally, a one-against-one directed acyclic graph multi-class support vector machine (OAO-DAG MSVM) was constructed and used to perform the classification. The results indicate that the image processing algorithm is fast and precise; the OAO-DAG MSVM gets a mean accuracy of 92.34% and a mean classification time of 12 ms, which can satisfy the accuracy and speed requirement of online classification of foreign fibers.

Local Wavelet Features for Statistical Object Classification and Localization

This article presents a system for texture-based probabilistic classification and localization of 3D objects in 2D digital images and discusses selected applications.

A Multimodal Constellation Model for Object Image Classification

We present an efficient method for object image classification. The method is an extention of the constellation model, which is a part-based model. Generally, constellation model has two weak points. (1) It is essentially a unimodal model which is unsuitable to be applied for categories with many types of appearances. (2) The probability function that represents the constellation model requires a high calculation cost.We introduced multimodalization and speed-up technique to the constellation model to overcome these weak points. The proposed model consists of multiple subordinate constellation models so that diverse types of appearances of an object category could be described by each of them, leading to the increase of description accuracy and consequently, improvement of the classification performance. In this paper, we present how to describe each type of appearance as a subordinate constellationmodel without any prior knowledge regarding the types of appearances, and also the implementation of the extended model's learning in realistic time. In experiments, we confirmed the effectiveness of the proposedmodel by comparison to methods using BoF, and also that the model learning could be realized in realistic time.

A method to incorporate uncertainty in the classification of remote sensing images

The aim of this paper is to investigate if the incorporation of the uncertainty associated with the classification of surface elements into the classification of landscape units (LUs) increases the results accuracy. To this end, a hybrid classification method is developed, including uncertainty information in the classification of very high spatial resolution multi-spectral satellite images, to obtain a map of LUs. The developed classification methodology includes the following steps: (1) a pixel-based hard classification with a probabilistic Bayesian classifier; (2) computation of the posterior probabilities and quantification of the classification uncertainty using an uncertainty measure; (3) image segmentation and (4) object classification based on decision rules. The classification of the resulting objects into LUs is performed considering a set of decision rules that incorporate the pixel-based classification uncertainty. The proposed methodology was tested on the classification of an IKONOS satellite image. The accuracy of the classification was computed using an error matrix. The comparison between the results obtained with the proposed approach and those obtained without considering the classification uncertainty revealed a 12% increase in the overall accuracy. This shows that the information about uncertainty can be valuable when making decisions and can actually increase the accuracy of the classification results.

A system for 3D texture-based probabilistic object recognition and its applications

This article presents a system for texture-based probabilistic classification and localisation of three-dimensional objects in two-dimensional digital images and discusses selected applications. In contrast to shape-based approaches, our texture-based method does not rely on object features extracted using image segmentation techniques. Rather, the objects are described by local feature vectors computed directly from image pixel values using the wavelet transform. Both gray level and colour images can be processed. In the training phase, object features are statistically modelled as normal density functions. In the recognition phase, the system classifies and localises objects in scenes with real heterogeneous backgrounds. Feature vectors are calculated and a maximisation algorithm compares the learned density functions with the extracted feature vectors and yields the classes and poses of objects found in the scene. Experiments carried out on a real dataset of over 40,000 images demonstrate the robustness of the system in terms of classification and localisation accuracy. Finally, two important real application scenarios are discussed, namely recognising museum exhibits from visitors' own photographs and classification of metallography images.