Segmentation Of Natural Scenes Research Articles

Understanding the time course and spatial biases of natural scene segmentation

Understanding the time course and spatial biases of natural scene segmentation

Semantic Segmentation of Surgical Instruments based on Enhanced Multi-scale Receptive Field

With the rapid development of robot assisted surgery, the segmentation of surgical instruments becomes more and more important. However, compared with the natural scene segmentation, surgical instrument segmentation is more difficult. To solve this problem, we improve a high and low resolution fusion module, which aims to extract detail information and context information from the fusion feature map of high and low resolution. Then, in the last layer of the encoder, we propose the Enhanced Multi-scale Receptive Field module to generate more available receptive fields. Our method is validated on 2017 MICCAI EndoVis Robotic Instrument Segmentation Challenge dataset, and the result is better than the other methods. The extended experiment is carried out on the dataset of our surgical soft robot which has a content implementation.

Multi scale mirror connection based encoder decoder network for text localization

Encoder decoder models with multi-scale feature concatenations have become ubiquitous for various natural scene segmentation tasks. In the current approach, a similar model with an improved mirror connection from encoders to decoder has been proposed. Three different types of mirror connections, namely, linear, parametric and convolutional, have been demonstrated in the proposed work. We have also implemented the use of internal skips to facilitate better gradient propagation within the encoder-decoder architecture. The proposed model also consists of an ensemble module that combines outputs from models with different kernel sizes, such as, 3 × 3, 5 × 5 and 7 × 7 to combine multi-scale features for efficient detections. The model was tested on the ICDAR 2003, SVT, ICDAR 2015 and the Total-Text dataset where it proved to be superior to other state of the art encoder-decoder architectures for pixel level classification.

BAS$^{4}$Net: Boundary-Aware Semi-Supervised Semantic Segmentation Network for Very High Resolution Remote Sensing Images

Semantic segmentation is a fundamental task in remote sensing image understanding. Recently, Deep Convolutional Neural Networks (DCNNs) have considerably improved the performance of the semantic segmentation of natural scenes. However, it is still challenging for very high-resolution remote sensing images. Due to the large and complex scenes as well as the influence of illumination and imaging angle, it is particularly difficult for the existing methods to accurately obtain the category of pixels at object boundaries—the so-called boundary blur. We propose a framework called Boundary-Aware Semi-Supervised Semantic Segmentation Network (BAS4Net), which obtains more accurate segmentation results without additional annotation workload, especially at the object boundaries. The Channel-weighted Multi-scale Feature (CMF) module balances semantic and spatial information, and the Boundary Attention Module (BAM) weights the features with rich semantic boundary information to alleviate the boundary blur. Additionally, to decrease the amount of difficult and tedious manual labeling of remote sensing images, a discriminator network infers pseudolabels from unlabeled images to assist semisupervised learning, and further improves the performance of the segmentation network. To validate the effectiveness of the proposed framework, extensive experiments have been performed on both the ISPRS Vaihingen dataset, and the novel remote sensing dataset AIR-SEG with more categories, and complex boundaries. The results demonstrate a significant improvement of accuracy especially on boundaries and for small objects.

Semantic Segmentation of Underwater Imagery Using Deep Networks Trained on Synthetic Imagery

Recent breakthroughs in the computer vision community have led to the emergence of efficient deep learning techniques for end-to-end segmentation of natural scenes. Underwater imaging stands to gain from these advances, however, deep learning methods require large annotated datasets for model training and these are typically unavailable for underwater imaging applications. This paper proposes the use of photorealistic synthetic imagery for training deep models that can be applied to interpret real-world underwater imagery. To demonstrate this concept, we look at the specific problem of biofouling detection on marine structures. A contemporary deep encoder–decoder network, termed SegNet, is trained using 2500 annotated synthetic images of size 960 × 540 pixels. The images were rendered in a virtual underwater environment under a wide variety of conditions and feature biofouling of various size, shape, and colour. Each rendered image has a corresponding ground truth per-pixel label map. Once trained on the synthetic imagery, SegNet is applied to segment new real-world images. The initial segmentation is refined using an iterative support vector machine (SVM) based post-processing algorithm. The proposed approach achieves a mean Intersection over Union (IoU) of 87% and a mean accuracy of 94% when tested on 32 frames extracted from two distinct real-world subsea inspection videos. Inference takes several seconds for a typical image.

Multi-nature hierarchical approach for natural image segmentation with pattern refinement feedback

A hierarchical learning method for segmenting natural images is proposed in this paper. This approach combines the perceptual information of three natures – colour, texture, and homogeneity – in order to segment natural colour images. These low-level features are extracted using a multiple scale neural architecture we previously proven in [1,20]. Present approach incorporates the human knowledge to a hierarchical categorisation process, where the features of the three natures are independently categorised. The final segmentation is achieved through pattern refinement cycles. The approach is compared to other two significant natural scene segmentation methods, achieving better results in a global evaluation. These comparisons are performed using the Berkeley Segmentation Dataset.

Adaptive perceptual color-texture image segmentation

We propose a new approach for image segmentation that is based on low-level features for color and texture. It is aimed at segmentation of natural scenes, in which the color and texture of each segment does not typically exhibit uniform statistical characteristics. The proposed approach combines knowledge of human perception with an understanding of signal characteristics in order to segment natural scenes into perceptually/semantically uniform regions. The proposed approach is based on two types of spatially adaptive low-level features. The first describes the local color composition in terms of spatially adaptive dominant colors, and the second describes the spatial characteristics of the grayscale component of the texture. Together, they provide a simple and effective characterization of texture that the proposed algorithm uses to obtain robust and, at the same time, accurate and precise segmentations. The resulting segmentations convey semantic information that can be used for content-based retrieval. The performance of the proposed algorithms is demonstrated in the domain of photographic images, including low-resolution, degraded, and compressed images.

Segmentation of natural scenes

A new simple and computationally efficient approach to image segmentation via recursive region splitting and merging is presented. Unlike other techniques the criterion for splitting is based on a generalization of a two-class gradient relaxation method and merging uses a test for mean gray level equivalency for adjacent regions. The technique is illustrated by providing results for both synthetic and natural scenes.

Extracting and Labeling Boundary Segments in Natural Scenes

This paper describes a set of algorithms used to perform segmentation of natural scenes through boundary analysis. The techniques include preprocessing, differentiation using a very simple operator, relaxation using case analysis, and postprocessing. The system extracts line segments as connected sets of edges, labels them, and computes features for them such as length and confidence.