Automatic Salient Object Extraction Based on Locally Adaptive Thresholding to Generate Tactile Graphics

Abstract

Automatic extraction of salient regions is beneficial for various computer vision applications, such as image segmentation and object recognition. The salient visual information across images is very useful and plays a significant role for the visually impaired in identifying tactile information. In this paper, we introduce a novel saliency cuts method using local adaptive thresholding to obtain four regions from a given saliency map. First, we produced four regions for image segmentation using a saliency map as an input image and local adaptive thresholding. Second, the four regions were used to initialize an iterative version of the GrabCuts algorithm and to produce a robust and high-quality binary mask with a full resolution. Finally, salient objects’ outer boundaries and inner edges were detected using the solution from our previous research. Experimental results showed that local adaptive thresholding using integral images can produce a more robust binary mask compared to the results from previous works that make use of global thresholding techniques for salient object segmentation. The proposed method can extract salient objects with a low-quality saliency map, achieving a promising performance compared to existing methods. The proposed method has advantages in extracting salient objects and generating simple, important edges from natural scene images efficiently for delivering visually salient information to the visually impaired.

A Systematic Literature Review on the Automatic Creation of Tactile Graphics for the Blind and Visually Impaired

“…JinSoo Cho and his research team presented an interesting project in the field of AT and tactile graphics generation for BVI individuals [31,54,58,59,66]. Meanwhile, Yoon et al [58] introduced a new method for extracting salient regions based on global contrast enhancement to improve the process of recognizing natural scene images for BVI individuals.…”

“…In the method, the contours of the salient regions were detected and translated into tactile graphics so that a BVI individual could perceive it using a refreshable braille device or printed swell paper. Abdusalomov et al [59] proposed a saliency cuts method using local adaptive thresholding to obtain four regions from a given saliency map [58]. In the method, the salient object is cut and the outer and inner edges of the salient objects are detected so that a BVI individual can effortlessly understand the content of an image.…”

“…To share multimedia content, they presented a braille electronic book reader application that could share a large number of figures, text, and audio content. Meanwhile, Tactile Pro and Tactile Edu mobile tablets and Braille Contents Author software were created for BVI individuals by combining the methods and technologies [10,31,54,58,59,66].…”

“…Travelling and tactile graphics 5 [35,36,[58][59][60] Map and audio guidance 3 [61][62][63] Artworks and object's shape 2 [64,65] Haptic and braille display 2 [31,66] Orientation and Mobility 1 [62] We identified types of input images to generate tactile graphics, as shown in Table 4. General images, such as natural scene images, were used the most (9 papers), followed by charts and diagrams (7 papers).…”

Improvement of the end-to-end scene text recognition method for “text-to-speech” conversion

“…11 Currently, with the universality of smart portable devices, there is a growing need for image/video-based applications, such as robot navigation, information retrieval from complex images, automatic reading of signboards and street signs, assistive navigation and geocoding. 1 Algorithms and methods for text detection and recognition are widely used in developed countries such as the United States, Germany, the United Kingdom, China, Japan, India and South Korea.…”