Local Neighborhood Intensity Pattern–A new texture feature descriptor for image retrieval

Abstract

In this paper, a new texture descriptor based on the local neighborhood intensity difference is proposed for content based image retrieval (CBIR). For computation of texture features like Local Binary Pattern (LBP), the center pixel in a 3×3 window of an image is compared with all the remaining neighbors, one pixel at a time to generate a binary bit pattern. It ignores the effect of the adjacent neighbors of a particular pixel for its binary encoding and also for texture description. The proposed method is based on the concept that neighbors of a particular pixel hold significant amount of texture information that can be considered for efficient texture representation for CBIR. The main impact of utilizing the mutual relationship among adjacent neighbors is that we do not rely on the sign of the intensity difference between central pixel and one of its neighbors ( ) only, rather we take into account the sign of difference values between and its adjacent neighbors along with the central pixels and same set of neighbors of . This makes our pattern more resistant to illumination changes. Moreover, most of the local patterns including LBP concentrates mainly on the sign information and thus ignores the magnitude. The magnitude information which plays an auxiliary role to supply complementary information of texture descriptor, is integrated in our approach by considering the mean of absolute deviation about each pixel from its adjacent neighbors. Taking this into account, we develop a new texture descriptor, named as Local Neighborhood Intensity Pattern (LNIP) which considers the relative intensity difference between a particular pixel and the center pixel by considering its adjacent neighbors and generate a sign and a magnitude pattern. Finally, the sign pattern and the magnitude pattern are concatenated into a single feature descriptor to generate a 2 more effective feature descriptor. The proposed descriptor has been tested for image retrieval on four databases, including three texture image databases -Brodatz texture image database, MIT VisTex database and Salzburg texture database and one face database -AT&T face database.The precision and recall values observed on these databases are compared with some state-of-art local patterns. The proposed method showed a significant improvement over many other existing methods.

Fractional Local Neighborhood Intensity Pattern for Image Retrieval using Genetic Algorithm

“…They considered the effect of two adjacent neighbors of a given neighboring pixel of the center. The work in [11] extended the idea using more adjacent neighbors and claimed a better performance. However, both [11] and [10] considered separate sign and magnitude patterns to encode each of them separately.…”

“…In our previous work on local neighborhood intensity pattern [11], we introduced the concept of local neighbors and generated a sign and magnitude pattern by considering the relative intensity difference between a particular pixel and the center pixel by considering its adjacent neighbors. In this work, we further extend the concept of exploiting texture information present in the local neighborhood of a particular pixel by considering the fractional change in the intensity in the local neighborhood and the center pixel of a 3×3 window with respect to a particular pixel and then comparing those values.…”

“…The descriptor named fractional local neighborhood intensity pattern is based on comparing the intensities of the center pixel of a 3×3 window with its neighboring pixels. It is inspired from the adjacent neighbor concept in LNIP [11]. In LNIP, it was shown that adjacent neighbors of a particular pixel contain useful texture information and it must be taken into account for calculating its relative intensity difference with respect to the center pixel of a 3x3 window rather than performing a direct comparison as used in LBP (Local Binary Pattern).…”

An Efficient Clustering Based CBIR System using Hadoop to Analyze Massive MRI Images Dataset for Early Disease Diagnosis

“…In this paper, 3D-LTCOP in combination with 3D-LTP is used for image retrieval purpose. Local Neighborhood Intensity Pattern (LNIP) is presented in [10] for pattern calculation. A new method based on orthogonal Fourier-Mellin moments (OFMMs) is proposed in [11] for effective indexing and retrieval of medical images.…”

A Texture Classification Approach Based on the Integrated Optimization for Parameters and Features of Gabor Filter via Hybrid Ant Lion Optimizer

“…Texture classification is one of the key problems in texture analysis, and it has been a long-standing research topic because of its importance in understanding the process of texture classification by humans and its extensive applications in computer vision and image analysis [3]. The main applications of texture classification include understanding medical images, extracting visible objects, retrieving content-based images, inspecting industrial faults [4][5][6][7], and so on.…”