Fruit Recognition Research Articles

The Use of the Combination of Texture, Color and Intensity Transformation Features for Segmentation in the Outdoors with Emphasis on Video Processing

Segmentation is the first and most important part in the development of any machine vision system with specific goals. Segmentation is especially important when the machine vision system works under environmental conditions, which means under natural light with natural backgrounds. In this case, segmentation will face many challenges, including the presence of various natural and artificial objects in the background and the lack of uniformity of light intensity in different parts of the camera's field of view. However, today, we must use different machine vision systems for outdoor use. For this reason, in this study, a segmentation algorithm was proposed for use in environmental conditions without the need for light control and the creation of artificial background using video processing with emphasizing the recognition of apple fruits on trees. Therefore, a video with more than 12 minutes duration containing more than 22,000 frames was studied under natural light and background conditions. Generally, in the proposed segmentation algorithm, five segmentation steps were used. These steps include: 1. Using a suitable color model; 2. Using the appropriate texture feature; 3. Using the intensity transformation method; 4. Using morphological operators; and 5. Using different color thresholds. The results showed that the segmentation algorithm had the total correct detection percentage of 99.013%. The highest sensitivity and specificity of segmentation algorithm were 99.224 and 99.458%, respectively. Finally, the results showed that the processor speed was about 0.825 seconds for segmentation of a frame.

Comparing bags of features, conventional convolutional neural network and AlexNet for fruit recognition

This paper presents a comparative study between Bag of Features (BoF), Conventional Convolutional Neural Network (CNN) and Alexnet for fruit recognition. Automatic fruit recognition can minimize human intervention in their fruit harvesting operations, operation time and harvesting cost. On the other hand, this task is very challenging because of the similarities in shapes, colours and textures among various types of fruits. Thus, a robust technique that can produce good result is necessary. Due to the outstanding performance of deep learning like CNN and its pre-trained models like AlexNet in image recognition, this paper investigates the accuracy of conventional CNN, and Alexnet in recognizing thirty different types of fruits from a publicly available dataset. Besides that, the recognition performance of BoF is also examined since it is one of the machine learning techniques that achieves good result in object recognition. The experimental results indicate that all of these three techniques produce excellent recognition accuracy. Furthermore, conventional CNN achieves the fastest recognition result compared to BoF, and Alexnet.

Boundary Value Analysis Testing on Augmented Reality of Indonesian Fruit Recognition at Mekarsari Tourist Park using Cloud Method on Android Mobile Devices

Software testing is a process of implementing a program with the aim of finding an error. A good test case is if the test has the possibility of finding an uncovered error. A successful test is if the test finds an error that was not initially found. One of the testing types available is black box testing. This paper proposes testing using black box testing technique. The black box testing method consists of several ways including equivalence partitioning, boundary value analysis, comparison testing, sample testing, robustness testing, and others. Among the many methods of testing, Boundary Value Analysis was chosen in this study. Boundary Value Analysis is a method of testing by determining the value of the lower limit and upper limit of the data that will be tested. This test is performed on the functions of Augmented Reality prototype of Indonesia fruit recognition by using the cloud method on Android mobile devices. From testing the distance of marker objects to Android mobile devices on cloud recognition using an Android camera shows that the higher the augmentable rating of the target image and the more the number of markers features detected, the easier the image will be tracked by AR. If the distance between the camera and the real object gets farther away, then the virtual object cannot be displayed. Testing with mobile devices using HSDPA, 3G and WIFI networks connecting a cloud database server to display virtual objects shows the average results of devices that use WIFI networks provide the fastest performance.

Apple Fruit Recognition Algorithm Based on Multi-Spectral Dynamic Image Analysis.

The ability to accurately recognize fruit on trees is a critical step in robotic harvesting. Many researchers have investigated a variety of image analysis methods based on different imaging technologies for fruit recognition. However, challenges still occur in the implementation of this goal due to various factors, especially variable light and proximal color background. In this study, images with fruit were acquired with a Forward Looking Infrared (FLIR) camera based on the Multi-Spectral Dynamic Imaging (MSX) technology. In view of its imaging mechanism, the optimal timing and shooting angle for image acquisition were pre-analyzed to obtain the maximum contrast between fruit and background. An effective algorithm was developed for locking potential fruit regions, which was based on the pseudo-color and texture information from MSX images. The algorithm was applied to 506 training and 340 evaluating images, including a variety of fruit and complex backgrounds. Recognition precision and sensitivity of these complete fruit regions were both above 92%, and those of incomplete fruit regions were not lower than 72%. The average processing time for each image was less than 1 s. The results indicated that the developed algorithm based on MSX imaging was effective for fruit recognition and could be suggested as a potential method for the automation of orchard production.

Pemanfaatan Augmented Reality Sebagai Media Pembelajaran Pengenalan Buah-Buahan

The concept of education for children is important. The aspects that must be considered are methods and learning media. In this research innovative and alternative learning media are made to understand fruits for children with Augmented Reality (AR). Augmented Reality (AR) in principle is a technology that is able to combine two-dimensional or three-dimensional virtual objects into a real environment and then project it. This learning media combines picture cards and virtual reality. Markers contained on picture cards will be captured by the mobile device camera, processed and will 3D animated pieces appear on the mobile screen in realtime. By using the concept of combining real world, real images on cards and virtual, applications can stimulate imagination and sense of desire in children and motivation to learn more and more. 3D fruit estimation created using the 3D Blender application and the Augmented Rea process lity is made using Unity and the Vuforia SDK library. The application of fruit recognition has been applied to several child respondents and has been tested on several types and brands of Android-based mobile phones. Based on research trials, 86% of 30 respondents stated that the application which was developed very effectively as a medium for the introduction of fruits.

Fast Recognition and Location of Target Fruit Based on Depth Information

In order to realize the recognition and localization of the target fruit more accurately and efficiently, an optimized graph-based recognition algorithm based on image depth information is proposed. Firstly, the gradient information of the depth image is obtained from the acquired depth apple image, the gradient vector is reflected from the 3D space to the 2D space, and then rotated 90 degrees clockwise, and the regular region forms the center of the vortex, which is the target fruit center to determine target fruit location. In the RGB space, the graph-based segmentation algorithm is introduced. By constructing the adaptive threshold, the mapping relationship between the number of centers and the parameters of k and min is established to find the target fruit region. Finally, scan the maximum radius of the super pixel area where the center of the circle is located, and fit the target fruit to realize the recognition and localization of the apple. The experimental results show that the target recognition rate is 96.61%, and the recognition time has been reduced significantly.

DURIAN RECOGNITION BASED ON MULTIPLE FEATURES AND LINEAR DISCRIMINANT ANALYSIS

Many fruit recognition approaches today are designed to classify different type of fruits but there is little effort being done for content-based fruit recognition specifically focuses on durian species. Durian, known as the king of tropical fruits, have few similar characteristics between different species where the skin have almost the same colour from green to yellowish brown with just slightly different shape and pattern of thorns. Therefore, it is hard to differentiate them with the current methods. It would be valuable to have an automated content-based recognition framework that can automatically represent and recognise a durian species given a durian image as the input. Therefore, this work aims to contribute to a new representation method based on multiple features for effective durian recognition. Two features based on shape and texture is considered in this work. Simple shape signatures which include area, perimeter, and circularity are used to determine the shape of the fruit durian and its base while the texture of the fruit is constructed based on Local Binary Pattern. We extracted these features from 240 durian images and trained this proposed method using few classifiers. Based on 10-fold cross validation, it is found that Logistic Regression, Gaussian Naive Bayesian, and Linear Discriminant Analysis classifiers performed equally well with 100% achievement of accuracy, precision, recall, and F1-score. We further tested the proposed algorithm on larger dataset which consisted of 42337 fruit images (64 various categories). Experimental results based on larger and more general dataset have shown that the proposed multiple features trained on Linear Discriminant Analysis classifier able to achieve 72.38% accuracy, 73% precision, 72% recall, and 72% F1-score.

Evaluation of CNN, Alexnet and GoogleNet for Fruit Recognition

Fruit recognition is useful for automatic fruit harvesting. Fruit recognition application can reduce or minimize human intervention during fruit harvesting operation. However, in computer vision, fruit recognition is very challenging because of similar shapes, colors and textures among various fruits. Illuminations changes due to weather condition also leads to a challenging task for fruit recognition. Thus, this paper tends to investigate the performance of basic Convolutional Neural Network (CNN), Alexnet and Googlenet in recognizing nine different types of fruits from a publicly available dataset. The experimental results indicate that all these techniques produce excellent recognition accuracy, but basic CNN achieves the fastest recognition result compared with Alexnet and Googlenet.

Recognising blueberry fruit of different maturity using histogram oriented gradients and colour features in outdoor scenes

For early yield estimation and harvest management, the recognition of blueberry fruit and their maturity is essential since blueberries do not ripen at the same time. This study was conducted to recognise visible blueberry fruit with different maturity using outdoor colour images acquired from a commercial field. The maturity of blueberry was divided into three different growth stages: mature, intermediate and young. The following stepwise algorithm was developed to identify the blueberry fruit: (1) A fruit training set was constructed using 1374 patches cropped from the original colour images. (2) HOG (Histogram Oriented Gradients) feature vectors were calculated from these patches, and a linear SVM (Support Vector Machine) classifier was trained to detect fruit-like regions rapidly. (3) Using a* and b* features in the L*a*b* colour space to discard non-fruit regions as well as categorise three fruit groups on those fruit-like regions. The KNN (K-nearest Neighbour) and a newly developed TMWE (Template Matching with Weighted Euclidean Distance) classifiers were applied to identify the fruit of different maturity. The performance of the method was evaluated using average detection accuracy on the testing images, missed rate, and incorrect detection rate of false positives. KNN classifier yielded the best average accuracy of 86.0%, 94.2% and 96.0% for young fruit, intermediate fruit and mature fruit, respectively. The proposed TMWE classifier gave a relatively high accuracy at lower computation cost. The results indicated that the method of this study is efficient in recognising blueberry fruit with different maturity using colour images in outdoor scenes.

Fruit recognition from images using deep learning

Abstract In this paper we introduce a new, high-quality, dataset of images containing fruits. We also present the results of some numerical experiment for training a neural network to detect fruits. We discuss the reason why we chose to use fruits in this project by proposing a few applications that could use such classifier.

A nearest neighbor approach for fruit recognition in RGB-D images based on detection of convex surfaces

Automatic fruit picking is a challenging problem in robotics with a wide application field. A prerequisite for realization of a robotic fruit picker is its ability to detect fruits in tree tops. An expert system, which would be able to compete with human perception, must be capable of recognizing fruits among leaves and branches under uncontrolled conditions, where fruits are occluded and shaded. In this paper, a novel approach for fruit recognition in RGB-D images based on detection and classification of convex surfaces is proposed. The input RGB-D image is first segmented into convex surfaces by a region growing procedure. Each convex surface is then described by an appropriate descriptor and classified with the aid of the associated descriptor. A novel descriptor of approximately convex surfaces is proposed, which we named Convex Template Instance (CTI) descriptor. It is based on approximating surfaces by convex polyhedrons with quantized face orientations, where every polyhedron face corresponds to one descriptor component. Computation of the proposed descriptor is simple and can be performed very efficiently. The proposed CTI descriptor is compared to the SHOT descriptor, a standard descriptor for 3D point clouds. Two variants of the both CTI and SHOT descriptor are evaluated, a variant which uses color and a variant which does not. A k-nearest neighbor classifier is used to classify detected surfaces into two classes: fruit and other. The main advantage of the proposed expert system in comparison to other fruit recognition solutions is its computational efficiency, which is of great importance for its target application – an automatic fruit picker. The proposed approach is evaluated using a challenging dataset containing RGB-D images of four fruit sorts acquired under uncontrolled conditions, which has been made publicly available to the scientific community, allowing benchmarking of novel fruit recognition methods.

Automatic Fruit Recognition Based on DCNN for Commercial Source Trace System

Automatic Fruit Recognition Based on DCNN for Commercial Source Trace System

Experiments and Analysis of Close-Shot Identification of On-Branch Citrus Fruit with RealSense.

Fruit recognition based on depth information has been a hot topic due to its advantages. However, the present equipment and methods cannot meet the requirements of rapid and reliable recognition and location of fruits in close shot for robot harvesting. To solve this problem, we propose a recognition algorithm for citrus fruit based on RealSense. This method effectively utilizes depth-point cloud data in a close-shot range of 160 mm and different geometric features of the fruit and leaf to recognize fruits with a intersection curve cut by the depth-sphere. Experiments with close-shot recognition of six varieties of fruit under different conditions were carried out. The detection rates of little occlusion and adhesion were from 80–100%. However, severe occlusion and adhesion still have a great influence on the overall success rate of on-branch fruits recognition, the rate being 63.8%. The size of the fruit has a more noticeable impact on the success rate of detection. Moreover, due to close-shot near-infrared detection, there was no obvious difference in recognition between bright and dark conditions. The advantages of close-shot limited target detection with RealSense, fast foreground and background removal and the simplicity of the algorithm with high precision may contribute to high real-time vision-servo operations of harvesting robots.

BLOB Analysis for Fruit Recognition and Detection

Robot application in agriculture can ease the farming process, especially as the harvesting robot for seasonal fruit that is available in a short time. The addition of "eye" as the image sensor is an important feature for a harvesting robot. Thanks to the increment of technology, the camera is getting smaller with better performance, and lower prices. The cheap sensors and components make the creation of cheap and effective robot possible. Image processing is necessary for object detection, and open source software is available now for this purpose. This paper proposes BLOB analysis for object detection of 5 fruits with different shapes and colors. The simulation results show that the proposed method is effective for object detection regardless the shapes, colors, and noises.

A Review of Computer Vision Methods for Fruit Recognition

A Review of Computer Vision Methods for Fruit Recognition

What is this fruit? Neural network application for Vietnamese fruit recognition

We use deep learning for problems in computer vision, image recognition and classification. Deep learning methods for fruit recognition are built with methods where features (in our case fruits key features) are processed and sent through multiple layers where transformations and computations are done sequentially to form a prediction model. Deep learning algorithms draws inspiration from many fields especially applied maths fundamentals like linear algebra, probability, information theory and numerical optimization. To the best of our knowledge this is the first web application for fruit recognition. Thanks to that users will be able to recognize most of the Vietnamese fruits without knowledge of Vietnamese language. What is more they can get a short description for each fruit and a video how to eat it. In the paper we compare different models of convolutional neural networks in order to find the best possible model of CNN. This system will is fine-tuned, what means that it learns on examples provided by users of application. Having that we propose algorithm which detects non-fruits pictures uploaded by user.

Research on Multi-class Fruits Recognition Based on Machine Vision and SVM

Picking fruit using by machine is an important research direction to liberate labor force and the first step of automatic picking is to recognize fruit. In order to improve the adaptability and reduce the cost of fruit picking robot, it is necessary to recognize multi-class fruit. In this paper, the recognition of multi-class fruit was studied with 6 kinds of fruit, such as apple, banana, citrus, carambola, pear, pitaya and so on. Firstly, the obtained fruit images were processed with Gaussian filter, histogram equalization and other image preprocessing. Secondly, the Otsu segmentation algorithm was used to segment the fruit image, and the edge of the image is extracted by the Canny edge detection operator. Thirdly, the shape invariant moment and other methods were used to synthesize the color and shape characteristics of the fruit to extract feature. Finally, the SVM classifier was applied to classify and recognize fruits according to the extracted feature vectors. The results showed that the recognition rate of 6 fruits, such as apples, bananas, citrus, carambola, pear and pitaya, were 95%, 80%, 97.5%, 86.7%, 92.5% and 96.7%, which could meet the needs of the fruit picking robot and lay the foundation for picking multi-class fruit by the picking robot.

The recognition of apple fruits in plastic bags based on block classification

The recognition of apple fruits in plastic bags is easy to be affected by reflected and refracted light. In order to weaken the influence of light, a method based on block classification is proposed. The method adopts watershed algorithm to segment original images into irregular blocks based on edge detection results of R–G grayscale images firstly. Compared with the watershed algorithm based on gradient images, the segmentation method can preserve fruits edge and reduce the number of blocks by 20.31%, because graying image method, R–G, filters most of leaves and edge detection operator insures that the edge of fruits are detected accurately. Next, these blocks are classified into fruit blocks and non-fruit blocks by support vector machine on the basis of the color and texture features extracted from blocks. Compared with the image recognition method based on pixel classification, the proposed method can restrain the interference of light caused by plastic bags effectively. The false negative rate (FNR) and false positive rate (FPR) of the method based on pixel classification are 21.71 and 14.53% respectively. The FNR and FPR of the proposed method are 4.65 and 3.50% respectively.

Multi-source data fusion using deep learning for smart refrigerators

Food recognition is one of the core functions for a smart refrigerator. But there are many challenges for accurate food recognition due to reasons of too many kinds of food inside the refrigerator which tends to obscure each other, and they may look very similar. This paper proposes a fruit recognition approach that fuses weight information and multi deep learning models. The proposed approach can remarkably improve recognition accuracy. We have extensively evaluated the proposed approach for its performance and accuracy, which demonstrate the effectiveness of the proposed approach.

Recognition and Matching of Clustered Mature Litchi Fruits Using Binocular Charge-Coupled Device (CCD) Color Cameras

Recognition and matching of litchi fruits are critical steps for litchi harvesting robots to successfully grasp litchi. However, due to the randomness of litchi growth, such as clustered growth with uncertain number of fruits and random occlusion by leaves, branches and other fruits, the recognition and matching of the fruit become a challenge. Therefore, this study firstly defined mature litchi fruit as three clustered categories. Then an approach for recognition and matching of clustered mature litchi fruit was developed based on litchi color images acquired by binocular charge-coupled device (CCD) color cameras. The approach mainly included three steps: (1) calibration of binocular color cameras and litchi image acquisition; (2) segmentation of litchi fruits using four kinds of supervised classifiers, and recognition of the pre-defined categories of clustered litchi fruit using a pixel threshold method; and (3) matching the recognized clustered fruit using a geometric center-based matching method. The experimental results showed that the proposed recognition method could be robust against the influences of varying illumination and occlusion conditions, and precisely recognize clustered litchi fruit. In the tested 432 clustered litchi fruits, the highest and lowest average recognition rates were 94.17% and 92.00% under sunny back-lighting and partial occlusion, and sunny front-lighting and non-occlusion conditions, respectively. From 50 pairs of tested images, the highest and lowest matching success rates were 97.37% and 91.96% under sunny back-lighting and non-occlusion, and sunny front-lighting and partial occlusion conditions, respectively.