Automatic Segmentation and Classification of Outdoor Images Using Neural Networks

1997 ◽  
Vol 08 (01) ◽  
pp. 137-144 ◽  
Author(s):  
N. W. Campbell ◽  
B. T. Thomas ◽  
T. Troscianko
Keyword(s):  
Neural Networks ◽  
Automatic Segmentation ◽  
Texture Features ◽  
Multi Layer Perceptron ◽  
Feature Map ◽  
Segmentation Process ◽  
Image Area

The paper describes how neural networks may be used to segment and label objects in images. A self-organising feature map is used for the segmentation phase, and we quantify the quality of the segmentations produced as well as the contribution made by colour and texture features. A multi-layer perceptron is trained to label the regions produced by the segmentation process. It is shown that 91.1% of the image area is correctly classified into one of eleven categories which include cars, houses, fences, roads, vegetation and sky.

Classification of Hepatic Lesions From CT Images Using Texture Features and Neural Networks

2001 ◽  
Author(s):  
M. Gletsos ◽  
S. G. Mougiakakou ◽  
G. K. Matsopoulos ◽  
K. S. Nikita ◽  
D. Kelekis
Keyword(s):  
Neural Networks ◽  
Texture Features ◽  
Ct Images ◽  
Hepatic Lesions

CLASSIFYING APPLES BY THE MEANS OF FLUORESCENCE IMAGING

2004 ◽  
Vol 18 (02) ◽  
pp. 157-174 ◽  
Author(s):  
MARIUS C. CODREA ◽  
OLLI S. NEVALAINEN ◽  
ESA TYYSTJÄRVI ◽  
MARTIN VANDEVEN ◽  
ROLAND VALCKE
Keyword(s):  
Neural Networks ◽  
Fluorescence Imaging ◽  
Chlorophyll A ◽  
Blue Light ◽  
Chlorophyll A Fluorescence ◽  
Classification Method ◽  
Test Material ◽  
Fluorescence Images

Classification of harvested apples when predicting their storage potential is an important task. This paper describes how chlorophyll a fluorescence images taken in blue light through a red filter, can be used to classify apples. In such an image, fluorescence appears as a relatively homogenous area broken by a number of small nonfluorescing spots, corresponding to normal corky tissue patches, lenticells, and to damaged areas that lower the quality of the apple. The damaged regions appear more longish, curved or boat-shaped compared to the roundish, regular lenticells. We propose an apple classification method that employs a hierarchy of two neural networks. The first network classifies each spot according to geometrical criteria and the second network uses this information together with global attributes to classify the apple. The system reached 95% accuracy using a test material classified by an expert for "bad" and "good" apples.

scholarly journals AQUiD: Automated Quality Assessment Using Digital Image Processing

2020 ◽  
pp. 295-300
Keyword(s):  
Oil Quality ◽  
Coconut Oil ◽  
Texture Features ◽  
High Quality ◽  
Quality Requirements ◽  
The People ◽  
Coconut Kernel ◽  
Growth Of Bacteria

Products in the market are expected to satisfy the consumer’s quality requirements. Agriculture being one of the main occupation of the people of India, the raw products must be sorted to determine whether they fit the quality description so that high quality products are obtained as the end result. The proposed method is designed to ensure the availability of good quality coconut oil in the market by assessing the quality of each individual sample going into the production line. 70% of moisture content present naturally in copra(dried coconut kernel) is dried to almost 7% for coconut oil production. To prevent the growth of bacteria and fungus on the surface of the copra, sulphur is added as a preservative. Allergenic reactions and lung performance restrictions can be caused due to the presence of sulphur in copra. The presence of moisture may also adversely affect oil quality. The texture features such as wrinkles, moulds, fungi growth on the surface also deplete the oil quality. The features of different kinds of copra are analysed and is used train the machine. The machine learning methodology is adopted for the classification of copra as usable and unusable.

scholarly journals Robust Skin lesion Classification via Machine Intelligence

2020 ◽  
Vol 8 (6) ◽  
pp. 4496-4500
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Skin Cancer ◽  
Texture Features ◽  
Machine Intelligence ◽  
Machine Learning Techniques ◽  
Learning Approaches ◽  
Image Size ◽  
Learning Techniques

Skin cancer is typically growth and spread of cells or lesion on the uppermost part or layer of skin known as the epidermis. It is one of rarest and deadliest found type of cancer, if undetected or untreated at early stages may lead in patient’s demise. Dermatologists use dermatoscopic images to identify the type of skin cancer by identification of asymmetry, border, colour, texture & size mole or a lesion. This method of detection can also be applied using machine learning techniques for classification these images into respective of cancer. There have been various studies and techniques which have been proposed various researchers across the globe in order to improve the classification of these dermatoscopic images. The proposed studies primarily focus on classification of dermatoscopic images based on lesion’s colour and texture features followed by intelligent machine learning approaches. Advances in these machine intelligent approaches such as deep neural networks and convolutional neural networks can be applied on dermatoscopic images to identify their features. A CNN based approach provides a additional accuracy over feature extraction as the algorithm is applied on pixel in overall image size. CNN also provides the ability to perform huge chunk of mathematical operations which is basic requirement in case of image processing and machine learning. The CNN based algorithm can be used to classify the dermatoscopic images with better efficiency and overall accuracy with having power of artificial-neural-network.

Sign in / Sign up

Export Citation Format

Share Document