Semantic Segmentation Network of Remote Sensing Images With Dynamic Loss Fusion Strategy

Abstract

The remote sensing (RS) images are widely used in various industries, among which semantic segmentation of RS images is a common research direction. At the same time, because of the complexity of target information and the high similarity of features between the classes, this task is very challenging. In recent years, semantic segmentation algorithms of RS images have emerged in an endless stream, but most of them are improved around the scale features of the target, and the accuracy has great room for improvement. In this case, we propose a semantic segmentation framework for RS images with dynamic perceptual loss. The framework is improved based on the InceptionV-4 network to form a network that includes contextual semantic fusion and dual-channel atrous spatial pyramid pooling (ASPP). The semantic segmentation network is an encoder-decoder structure. In addition, we design a dynamic perceptual loss module and a dynamic loss fusion strategy by further observing the loss changes of the network, so as to better improve the classified details. Finally, experiment on the ISPRS 2D Semantic Labeling Contest Vaihingen Dataset and Massachusetts Building Dataset. Compared with some segmentation networks, our model has excellent performance.

Research Contribution and Comprehensive Review towards the Semantic Segmentation of Aerial Images Using Deep Learning Techniques

“…In 2021, Jiang [80] has suggested a semantic segmentation model using high-resolution remote-sensing images through CNN and mask generation, in which the NN architecture was intended for obtaining a precise mask. In 2021, Liu et al [81] have designed a new semantic segmentation model using remote-sensing images using Inceptionv-4 network for getting the enhanced classified information. is model has introduced the fusion of features for solving the classification of edge of objects.…”

“…A general scheme for constructing a deep network to process a rich dataset is complex. e improved DNN models are modified inceptionV-4 network [50], NDRB [52], ResNet101-v2 [39], ALRNet [70], HMANET [56], ResNet [65], inceptionV-4 network [81], MAVNet [41], and SDNF [61], which are aimed to enhance the superior accuracy on segmentation.…”

“…Modified InceptionV-4 Network [50] Nested Dilated Residual Block (NDRB) [52] ResNet101-V2 [39] ALRNet [70] HMANET [56] ResNet [65] Inception V-4 Network [81] MAVNet [41] SDNF [61] Segment Connection Algorithm [39] DSMFNets [40] Hard Sampling and Soft Sampling [74] Aerial-BiSeNet [75] MAML-PSO [63] DASSN-RSI [79] Cascaded Framework [34] Figure 3: Algorithmic categorization of existing semantic segmentation models. Although the potsdam dataset has offered consistent performance, there is a considerable reduction in the vaihingen IR-R-G dataset concerning IOU scores [51] Relation module-equipped FCN e performance of semantic segmentation is enhanced with the use of a network using aerial scenes However, the suggested relation modules regarding segmentation are basic one, and thus, it does not offer superior efficiency [52] Nested dilated residual block (NDRB)…”

“…Citation number Dataset description [35] Vaihingen and potsdam [50] International society for photogrammetry and remote-sensing (ISPRS) 2D semantic labeling contest potsdam and inria aerial image labeling dataset [28] ISPRS [42] Manual dataset on Tokyo [72] Big data and computing intelligence contest (BDCI) [54] Cityscapes [58] e remote-sensing data collected from field 1 [62] ISPRS 2D semantic labeling benchmark dataset [63] 2015 igrss data fusion competition [32] Rit-18 [79] Gaofen image dataset (GID) datasets [80] Potsdam and vaihingen datasets [67] Luxcarta dataset [81] ISPRS 2D semantic labeling contest vaihingen dataset and Massachusetts building dataset [68] Ottawa dataset, stone gate dataset, sardinia dataset, yellow river estuary dataset, barbara dataset, and USGS dataset [49] Landsat-8 satellite and ISPRS vaihingen challenge dataset [83] UCM dataset and the deepglobe dataset samples, uncommonly the ones which manage little datasets such as in the medical domain. Augmentation of data serves to expand the quantity of preparing tests by applying a set of changes either in the information space, or element space, or now and again both to the images, i.e., both the input image and the segmentation map.…”

Land Cover Classification of Resources Survey Remote Sensing Images Based on Segmentation Model

“…From the perspective of published articles in recent years, the research focus of scholars can be divided into two aspects as a whole. Many scholars try to improve the ability of classification, segmentation, or detection of the DL algorithm itself, to improve the accuracy of multi-objective classification, improve the effect of detail resolution, solve the problems of insufficient training samples, unbalanced labeling samples, and so on [14], [15]. At present, it is popular to improve based on classical convolutional neural network (CNN) and include the information processing skills of attention mechanism in the design [16], [17].…”