Automatic Defect Segmentation on Leather with Deep Learning

Abstract

Leather is a natural and durable material created through a process of tanning of hides and skins of animals. The price of the leather is subjective as it is highly sensitive to its quality and surface defects condition. In the literature, there are very few works investigating on the defects detection for leather using automatic image processing techniques. The manual defect inspection process is essential in an leather production industry to control the quality of the finished products. However, it is tedious, as it is labour intensive, time consuming, causes eye fatigue and often prone to human error. In this paper, a fully automatic defect detection and marking system on a calf leather is proposed. The proposed system consists of a piece of leather, LED light, high resolution camera and a robot arm. Succinctly, a machine vision method is presented to identify the position of the defects on the leather using a deep learning architecture. Then, a series of processes are conducted to predict the defect instances, including elicitation of the leather images with a robot arm, train and test the images using a deep learning architecture and determination of the boundary of the defects using mathematical derivation of the geometry. Note that, all the processes do not involve human intervention, except for the defect ground truths construction stage. The proposed algorithm is capable to exhibit 91.5% segmentation accuracy on the train data and 70.35% on the test data. We also report confusion matrix, F1-score, precision and specificity, sensitivity performance metrics to further verify the effectiveness of the proposed approach.

Efficient Neural Network Approaches for Leather Defect Classification

“…Our previous work [11] discusses the elicitation of the dataset in detail and locates the fly bite defect with a segmentation accuracy of 91%. The experiment is examined on a relatively small dataset that only contains 584 images.…”

“…The sample images are shown in Figure 4. More information about the elicitation of the dataset can be found in [11,12].…”

Assessment of Convolutional Neural Network Pre-Trained Models for Detection and Orientation of Cracks

“…These infrastructures are constantly under stress due to natural and man-made hazards, such as earthquakes, blasts, and daily use. These stresses cause a variety of deteriorations, one of which is crack development [ 3 , 4 ]. Automation detection of these defects can significantly reduce the time and cost associated with inspection.…”

Integrated Neural Network and Machine Vision Approach For Leather Defect Classification

“…In a recent work on automated defect segmentation, Liong et al [6] introduced a series of algorithms to predict tick-bite defects on leather samples . Different from the conventional methods that capture the leather sample manually, [6] elicits all the image data using a robot arm and draws the defect region with a chalk using the same robot arm.…”

“…In a recent work on automated defect segmentation, Liong et al [6] introduced a series of algorithms to predict tick-bite defects on leather samples . Different from the conventional methods that capture the leather sample manually, [6] elicits all the image data using a robot arm and draws the defect region with a chalk using the same robot arm. One of the state-of-the-art models for instance segmentation, namely, Mask Region-based Convolutional Neural Network (Mask R-CNN) is employed and fine tuned with 84 defective images to learn the local features of the leather samples.…”