For effectness verification of our proposed neural network, a total of 19,368 lab-made images of butterfly specimensspanning 48 sub-species areutilizedas testing samples, while 116,208 augmented images are employed for training.
This dataset contains laser scans of PCBs as explained in "Fault Diagnosis in Microelectronics Attachment via Deep Learning Analysis of 3D Laser Scans". On the left and right image, we have a closer look at one circuit module of a PCB , before and after die attachment. Notice the different types of glue annotated as A, B, C, D and E. On each circuit there are four glue deposits on each type where approximately the same quantity of glue has been placed. As explainedin our paper the top three deposits are used for training and the bottom one for testing.
The date fruit dataset was created to address the requirements of many applications in the pre-harvesting and harvesting stages. The two most important applications are automatic harvesting and visual yield estimation. The dataset is divided into two subsets and each of them is oriented into one of these two applications. The first dataset consists of 8079 images of more than 350 date bunches captured from 29 date palms. The date bunches belong to five date types: Naboot Saif, Khalas, Barhi, Meneifi, and Sullaj.
Please refer to the readme and documentation files for a full description.
The complete documentation of the dataset is available in the following article:
Some experiments were performed on the dataset-1 and the results can be found in the following article: [To download the dataset-1 as processed and categorized in this article, refer to the zip file named "DATASET-1 (224 X 224) Categorized.zip"]
You can visit our website for more details and related resources:
This data is divided into two sets.
DATASET-1 consists of 8079 images of date bunches that were taken using two color cameras from different angles and scales during one season in six imaging sessions (recording times) over the period of Jun-Sep 2016. The date bunches belong to five date varieties: Naboot Saif, Khalas, Barhi, Meneifi, and Sullaj.
DATASET-1 zip files:
---- DATASET-1.zip (42 GB)
High resolution images (8079 images)
---- DATASET-1_(224 X 224).zip (108 MB)
A preview of the 8079 images. Images size: 224 X 224.
---- DATASET-1 (224 X 224) Categorized.zip (649 MB)
This file contains the images of dataset-1 resized to 224x224 pixels and categorized into subfolders according to date fruit type, maturity state, and the harvesting decision. This categorization, and image resolution, is related to the experiments described in ref .
---- DATASET-1_(annotation files).zip (104 KB)
The annotation (labeling) files for type classification, maturity analysis, and harvesting decision applications. The labeling instructions and rules are explained in detail in Ref. .
DATASET-2 contains images, videos, and weight measurements that were acquired during the harvesting period of Barhi dates.
---- DATASET-2_images_Barhi_date_bunches_on_orchard.zip (16.44 GB)
Images of 152 Barhi date bunches before and during harvesting.
---- DATASET-2_images_Barhi_date_bunches_front_graph_paper.zip (88.16 MB)
Images of the 152 Barhi date bunches in front of graph paper.
---- DATASET-2_weight_measurements_Barhi_date_bunches.xlsx (683.99 KB)
The weight measurements of the 152 Barhi date bunches.
---- DATASET-2_videos_Barhi_date_palms.zip (4.49 GB)
360-degree videos of the nine Barhi date palms (120 date bunches).
---- DATASET-2_images&measurements_Sullaj_date_bunches_front_graph_paper.zip (76 MB)
Images of 11 Sullaj date bunches in front of graph paper with weight & dimensions measurements.
---- DATASET-2_images&measurements_individual_dates&bunches_stalks.zip (549.18 MB)
Images of individual dates and bunches' stalks with weight & dimensions measurements.
Occlusion, glare and secondary reflections formed due to and on the spectacles - results in poor detection, localization, and recognition of eye/face features. We term all the problems related to the usage of spectacles as The spectacle problem. Though several studies on the spectacle detection and removal have been reported in the literature, the study focusing on spectacle problem removal is very limited. One of the main reasons being, the nonavailability of a facial image database highlighting the spectacle problems.