Hand-crafted Methods Research Articles

ARACAM: A RGB-D Multi-View Photogrammetry System for Lower Limb 3D Reconstruction Applications.

In the world, there is a growing need for lower limb prostheses due to a rising number of amputations caused primarily, by diabetic foot. Researchers enable functional and comfortable prostheses through prosthetic design by integrating new technologies applied to the traditional handcrafted method for prosthesis fabrication that is still current. That is why computer vision shows to be a promising tool for the integration of 3D reconstruction that may be useful for prosthetic design. This work has the objective to design, prototype, and test a functional system to scan plaster cast molds, which may serve as a platform for future technologies for lower limb reconstruction applications. The image capture system comprises 5 stereoscopic color and depth cameras, each with 4 DOF mountings on an enveloping frame, as well as algorithms for calibration, segmentation, registration, and surface reconstruction. The segmentation metrics of dice coefficient and Hausdorff distance (HD) show strong visual similarity with an average similarity of 87% and average error of 6.40 mm, respectively. Moving forward, the system was tested on a known 3D printed model obtained from a computer tomography scan to which comparison results via HD show an average error of ≤1.93 mm thereby making the system competitive against the systems reviewed from the state-of-the-art.

IMAGE ORIENTATION WITH A HYBRID PIPELINE ROBUST TO ROTATIONS AND WIDE-BASELINES

Abstract. The extraction of reliable and repeatable interest points among images is a fundamental step for automatic image orientation (Structure-From-Motion). Despite recent progresses, open issues in challenging conditions - such as wide baselines and strong light variations - are still present. Over the years, traditional hand-crafted methods have been paired by learning-based approaches, progressively updating the state-of-the-art according to recent benchmarks. Notwithstanding these advancements, learning-based methods are often not suitable for real photogrammetric surveys due to their lack of rotation invariance, a fundamental requirement for these specific applications. This paper proposes a novel hybrid image matching pipeline which employs both hand-crafted and deep-based components, to extract reliable rotational invariant keypoints optimized for wide-baseline scenarios. The proposed hybrid pipeline was compared with other hand-crafted and learning-based state-of-the-art approaches on some photogrammetric datasets using metric ground-truth data. Results show that the proposed hybrid matching pipeline has high accuracy and appeared to be the only method among the evaluated ones able to register images in the most challenging wide-baseline scenarios.

A Comparative Study: Toward an Effective Convolutional Neural Network Architecture for Sensor-Based Human Activity Recognition

The feature extraction of human activity recognition (HAR) based on sensor data has been studied as a hand-crafted method. The significant feature extraction ability is a key factor in improving the accuracy of HAR. Recently, deep learning methods have been employed for feature extraction. In this paper, we review previous studies on deep learning methods in HAR and discuss suitable models for feature extraction. First, we applied various convolutional neural networks to clarify the effective architecture for HAR. Afterward, we developed advanced models by embedding submodules, such as self-attention and recurrent neural networks, often adopted in recent studies. Comparative experiments on HASC, UCI, and WISDM public datasets showed that Inception-V3, which used cross-channel multi-size convolution transformation, outperformed other backbones. Through comparative experiments after embedding submodules, submodules do not always have a positive effect on accuracy. Compared with other submodules, SENet has a positive effect. We conclude that it is essential to select an appropriate backbone model before applying the submodules, and submodules are unnecessary in some cases.

Pseudo Complex-Valued Deformable ConvLSTM Neural Network With Mutual Attention Learning for Hyperspectral Image Classification

Convolutional long short-term memory (ConvLSTM) has received much attention for hyperspectral image (HSI) classification due to its ability of modeling long-range correlations, which, however, is vulnerable to too many parameters and insufficient training, limiting its classification accuracy, especially for small samples. Different from it, traditional hand-crafted methods extract the features with basic attributes of HSIs, which can provide the lack of details and interpretability of deep semantic features. However, existing methods fail to incorporate their complementarity for HSI classification. As such, a Pseudo complex-valued (CV) Deformable ConvLSTM Neural Network with mutual Attention learning (APDCLNN) is proposed, providing a new way to realize the collaborative learning of hand-crafted and deep features for HSI classification. First, a 2-D pseudo CV deformable ConvLSTM (PDConvLSTM2D) cell is designed using deformable convolution and complex operations, with which a spatial–spectral PDConvLSTM2D neural network (SSPDCL2DNN) is built to extract scale- and spectral-enhanced deep spatial–spectral features. Then, 3-D Gabor filter is used to extract hand-crafted features, and a mutual attention-based multimodality feature learning and fusion (MAMLF) module is designed to integrate them into deep features for training and optimization of SSPDCL2DNN. Finally, an attention loss subnetwork is designed to refine the classification results. As we know, this is the first attempt to apply the idea of mutual attention learning to fuse hand-crafted and deep features for HSI classification. Extensive experiments on three widely used HSI datasets show the advantages of our model over other deep methods in terms of both quantitative and visual quality.

A new function of stereo matching algorithm based on hybrid convolutional neural network

This paper proposes a new hybrid method between the learning-based and handcrafted methods for a stereo matching algorithm. The main purpose of the stereo matching algorithm is to produce a disparity map. This map is essential for many applications, including three-dimensional (3D) reconstruction. The raw disparity map computed by a convolutional neural network (CNN) is still prone to errors in the low texture region. The algorithm is set to improve the matching cost computation stage with hybrid CNN-based combined with truncated directional intensity computation. The difference in truncated directional intensity value is employed to decrease radiometric errors. The proposed method’s raw matching cost went through the cost aggregation step using the bilateral filter (BF) to improve accuracy. The winner-take-all (WTA) optimization uses the aggregated cost volume to produce an initial disparity map. Finally, a series of refinement processes enhance the initial disparity map for a more accurate final disparity map. This paper verified the performance of the algorithm using the Middlebury online stereo benchmarking system. The proposed algorithm achieves the objective of generating a more accurate and smooth disparity map with different depths at low texture regions through better matching cost quality.

Pigment, light, paper, screen; formats as interacting multi-dimensional creative works

Formats have an interacting range of qualities that practice based research can explore, so testing and questioning theory; they are more than dimensions. Video art can play with space, light and sound, through a choice of materials, including visuals and audio. I make video art out of the virtual world of Second Life with its interplay of two-dimensional textures and three-dimensional objects. Editing this virtual world digital video with digitised archive film, contemporary music and literature explores how the formats used by multiple creators across time can interact, so exploring meanings through reinterpretation and translation. Taking still images from that video art into printmaking using photogravure etching, a hand-crafted method of mechanically pressing ink on paper, considers the translation of light on screen to ink on paper, and explores format from another position. Going from moving image to still images changes the relationship with time and distance, and from ratios in pixels to dimensions in millimetres.

Vision Transformer-Based Tailing Detection in Videos

Tailing is defined as an event where a suspicious person follows someone closely. We define the problem of tailing detection from videos as an anomaly detection problem, where the goal is to find abnormalities in the walking pattern of the pedestrians (victim and follower). We, therefore, propose a modified Time-Series Vision Transformer (TSViT), a method for anomaly detection in video, specifically for tailing detection with a small dataset. We introduce an effective way to train TSViT with a small dataset by regularizing the prediction model. To do so, we first encode the spatial information of the pedestrians into 2D patterns and then pass them as tokens to the TSViT. Through a series of experiments, we show that the tailing detection on a small dataset using TSViT outperforms popular CNN-based architectures, as the CNN architectures tend to overfit with a small dataset of time-series images. We also show that when using time-series images, the performance of CNN-based architecture gradually drops, as the network depth is increased, to increase its capacity. On the other hand, a decreasing number of heads in Vision Transformer architecture shows good performance on time-series images, and the performance is further increased as the input resolution of the images is increased. Experimental results demonstrate that the TSViT performs better than the handcrafted rule-based method and CNN-based method for tailing detection. TSViT can be used in many applications for video anomaly detection, even with a small dataset.

Towards real time guide wire shape extraction in fluoroscopic sequences: A two phase deep learning scheme to extract sparse curvilinear structures.

Real time localization and shape extraction of guide wire in fluoroscopic images plays a significant role in the image guided navigation during cerebral and cardiovascular interventions. Given the complexity of the non-rigid and sparse characteristics of guide wire structures, and the low SNR(Signal Noise Ratio) of fluoroscopic images, traditional handcrafted guide wire tracking methods such as Frangi filter, Hessian Matrix, or open active contour usually produce insufficient accuracy with high computational cost, and may require extra human intervention for proper initialization or correction. The application of deep learning techniques to guide wire tracking is reported to produce significant improvement in guide wire localization accuracy, but the heavy calculation cost is still a concern. In this paper we propose a two phase deep learning scheme for accurate and real time guide wire shape extraction in fluoroscopic sequences. In the first phase we train a guide wire localization network to pick image regions containing guide wire structures. From the picked image regions, we train a guide wire shape extraction network in the second phase to mark the guide wire pixels. We report that our proposed method can accurately distinguish about 99% of the guide wire structure pixels, and the falsely detected pixels in the background are close to 0, the average offset from the ground truth is less than 1 pixel. For extreme cases where traditional handcrafted method may fail, our proposed method can still extract guide wire completely and accurately. The processing time for a 512 × 512 pixels image is 78 ms. Compared with the traditional filtering based method from our previous work, we show that our proposed method can achieve more accurate and stable performance. Compared with other deep learning methods, our proposed method significantly improve calculation efficiency to meet the real time requirement of clinical applications.

Iris Liveness Detection for Biometric Authentication: A Systematic Literature Review and Future Directions

Biometrics is progressively becoming vital due to vulnerabilities of traditional security systems leading to frequent security breaches. Biometrics is an automated device that studies human beings’ physiological and behavioral features for their unique classification. Iris-based authentication offers stronger, unique, and contactless identification of the user. Iris liveness detection (ILD) confronts challenges such as spoofing attacks with contact lenses, replayed video, and print attacks, etc. Many researchers focus on ILD to guard the biometric system from attack. Hence, it is vital to study the prevailing research explicitly associated with the ILD to address how developing technologies can offer resolutions to lessen the evolving threats. An exhaustive survey of papers on the biometric ILD was performed by searching the most applicable digital libraries. Papers were filtered based on the predefined inclusion and exclusion criteria. Thematic analysis was performed for scrutinizing the data extracted from the selected papers. The exhaustive review now outlines the different feature extraction techniques, classifiers, datasets and presents their critical evaluation. Importantly, the study also discusses the projects, research works for detecting the iris spoofing attacks. The work then realizes in the discovery of the research gaps and challenges in the field of ILD. Many works were restricted to handcrafted methods of feature extraction, which are confronted with bigger feature sizes. The study discloses that dep learning based automated ILD techniques shows higher potential than machine learning techniques. Acquiring an ILD dataset that addresses all the common Iris spoofing attacks is also a need of the time. The survey, thus, opens practical challenges in the field of ILD from data collection to liveness detection and encourage future research.

Identification of glioblastoma molecular subtype and prognosis based on deep MRI features

Magnetic resonance imaging (MRI) has become an important tool to study the correlation between the imaging phenotypes and the molecular profiles of Glioblastoma multiforme (GBM), the most frequent and lethal brain tumor. This type of study is named “Radiogenomics”. Currently, many radiogenomics studies segmented the tumors manually and then extracted hand-crafted MRI features for analysis. Automated segmentation approach as well as automatically learned features are urgently needed to release the burden of manual operation. In this study, we developed a predictive model, named DeepRA, based on deep imaging features and machine learning technologies to identify MRI signatures that enable accurate prediction of GBM molecular subtype and patient overall survival. We here for the first time used state-of-the-art deep imaging features to predict both molecular subtype and overall survival. We converted the high-dimensional deep feature representations to interpretable feature vector, selected the most distinguishing features and conducted the prediction. Experiments validated on The Cancer Genome Atlas (TCGA) data have shown that the DeepRA outperformed the traditional hand-crafted method. Also, compared with the regular convolutional neural networks used to segment tumors, the DeepRA presents a better performance, which shows the features extracted from DeepRA are more predictive. The implementation of the proposed method is available at https://github.com/RanSuLab/GBM-subtype-survival.

EvoStencils: a grammar-based genetic programming approach for constructing efficient geometric multigrid methods

For many systems of linear equations that arise from the discretization of partial differential equations, the construction of an efficient multigrid solver is challenging. Here we present EvoStencils, a novel approach for optimizing geometric multigrid methods with grammar-guided genetic programming, a stochastic program optimization technique inspired by the principle of natural evolution. A multigrid solver is represented as a tree of mathematical expressions that we generate based on a formal grammar. The quality of each solver is evaluated in terms of convergence and compute performance by automatically generating an optimized implementation using code generation that is then executed on the target platform to measure all relevant performance metrics. Based on this, a multi-objective optimization is performed using a non-dominated sorting-based selection. To evaluate a large number of solvers in parallel, they are distributed to multiple compute nodes. We demonstrate the effectiveness of our implementation by constructing geometric multigrid solvers that are able to outperform hand-crafted methods for Poisson’s equation and a linear elastic boundary value problem with up to 16 million unknowns on multi-core processors with Ivy Bridge and Broadwell microarchitecture.

P–141 Artificial intelligence system for the automation of the blastocyst morphology evaluation in GERI Time-lapse Incubator

Abstract Study question Can an Artificial Intelligence (AI) system (hand-crafted vs. deep learning techniques) based on single embryo image analysis from a GERI time-lapse incubator (TL) evaluate the blastocyst morphology? Summary answer Our hand-crafted method trained with blastocyst images from Geri-TL evaluated and classified parameters regarding to embryo quality with a global precision of 63.7% in blind-test. What is known already Recent studies have shown that AI can improve automatic grading and embryo selection. The approaches that have been carried out are very different, but all they conclude that there is a great potential (Rad2019, Manoj2020, Thirumalaraju2020). As we know, conventional embryo evaluation is performed manually based on the morphology of the blastocyst, therefore, it should be possible to replicate this process. In this study, we implemented different methods to analyse the behaviour and performance of an AI doing embryology tasks. Study design, size, duration Our study consisted of a retrospective analysis for the automatization of embryo evaluation with different approaches. We developed our models based on 715 images extracted from GERI TL Videos (Genea, Australia) from a single IVF center. Database was divided into 3 classes depending on the quality of the embryo according to ASEBIR morphology criteria (high; medium and low-quality). All the images were divided into 70% for training, 15% for validating and 15% for testing. Participants/materials, setting, methods We developed an automated AI algorithm to extract and classify features from images at 111,5 hpi of embryos cultured in GERI TL. Hand-crafted features from texture information are extracted to feed the classification algorithm. A statistical analysis is carried out to select the more discriminative variables. Parallelly, a deep neural network was built to compare performance of automatic and hand-crafted features. Additionally, we trained a model to detect embryo in the well. Main results and the role of chance High-quality, medium-quality and low-quality sensitivity were 73%, 56% and 72% for hand-crafted method and 76%, 53% and 22% for deep learning approach, respectively. High-quality, medium-quality and low-quality precision were 66%, 56% and 76% for hand-crafted method and 40%, 60% and 55% for deep learning approach, respectively. The global accuracy associated with each method was 64% and 50%. Also, we noticed that results were higher when we applied our embryo masks that avoid irrelevant information. In this initial attempt, our results showed that it is possible to replicate the embryo evaluation process. Limitations, reasons for caution The low results obtained in our deep learning model due to the absence of an extent dataset did not allow to obtain a model applicable to the clinic. However, the preliminary study let us to conclude the high potential of the approach. Wider implications of the findings: Our results showed a potential automatization of the embryo evaluation process in Geri TL where the available software for embryo selection does not provide such option. Our findings leaded to an increase in objectification, a reduction of the workload of the embryologist and the research of new unknown morphological variables. Trial registration number Not applicable

Automated segmentation of macular edema for the diagnosis of ocular disease using deep learning method

Macular edema is considered as a major cause of visual loss and blindness in patients with ocular fundus diseases. Optical coherence tomography (OCT) is a non-invasive imaging technique, which has been widely applied for diagnosing macular edema due to its non-invasive and high resolution properties. However, the practical applications remain challenges due to the distorted retinal morphology and blurred boundaries near macular edema. Herein, we developed a novel deep learning model for the segmentation of macular edema in OCT images based on DeepLab framework (OCT-DeepLab). In this model, we used atrous spatial pyramid pooling (ASPP) to detect macular edema at multiple features and used the fully connected conditional random field (CRF) to refine the boundary of macular edema. OCT-DeepLab model was compared against the traditional hand-crafted methods (C-V and SBG) and the end-to-end methods (FCN, PSPnet, and U-net) to estimate the segmentation performance. OCT-DeepLab showed great advantage over the hand-crafted methods (C-V and SBG) and end-to-end methods (FCN, PSPnet, and U-net) as shown by higher precision, sensitivity, specificity, and F1-score. The segmentation performance of OCT-DeepLab was comparable to that of manual label, with an average area under the curve (AUC) of 0.963, which was superior to other end-to-end methods (FCN, PSPnet, and U-net). Collectively, OCT-DeepLab model is suitable for the segmentation of macular edema and assist ophthalmologists in the management of ocular disease.

Convolutional Neural Network with an Elastic Matching Mechanism for Time Series Classification

Recently, some researchers adopted the convolutional neural network (CNN) for time series classification (TSC) and have achieved better performance than most hand-crafted methods in the University of California, Riverside (UCR) archive. The secret to the success of the CNN is weight sharing, which is robust to the global translation of the time series. However, global translation invariance is not the only case considered for TSC. Temporal distortion is another common phenomenon besides global translation in time series. The scale and phase changes due to temporal distortion bring significant challenges to TSC, which is out of the scope of conventional CNNs. In this paper, a CNN architecture with an elastic matching mechanism, which is named Elastic Matching CNN (short for EM-CNN), is proposed to address this challenge. Compared with the conventional CNN, EM-CNN allows local time shifting between the time series and convolutional kernels, and a matching matrix is exploited to learn the nonlinear alignment between time series and convolutional kernels of the CNN. Several EM-CNN models are proposed in this paper based on diverse CNN models. The results for 85 UCR datasets demonstrate that the elastic matching mechanism effectively improves CNN performance.

Attention Multi-Scale Network for Automatic Layer Extraction of Ice Radar Topological Sequences

Analyzing the surface and bedrock locations in radar imagery enables the computation of ice sheet thickness, which is important for the study of ice sheets, their volume and how they may contribute to global climate change. However, the traditional handcrafted methods cannot quickly provide quantitative, objective and reliable extraction of information from radargrams. Most traditional handcrafted methods, designed to detect ice-surface and ice-bed layers from ice sheet radargrams, require complex human involvement and are difficult to apply to large datasets, while deep learning methods can obtain better results in a generalized way. In this study, an end-to-end multi-scale attention network (MsANet) is proposed to realize the estimation and reconstruction of layers in sequences of ice sheet radar tomographic images. First, we use an improved 3D convolutional network, C3D-M, whose first full connection layer is replaced by a convolution unit to better maintain the spatial relativity of ice layer features, as the backbone. Then, an adjustable multi-scale module uses different scale filters to learn scale information to enhance the feature extraction capabilities of the network. Finally, an attention module extended to 3D space removes a redundant bottleneck unit to better fuse and refine ice layer features. Radar sequential images collected by the Center of Remote Sensing of Ice Sheets in 2014 are used as training and testing data. Compared with state-of-the-art deep learning methods, the MsANet shows a 10% reduction (2.14 pixels) on the measurement of average mean absolute column-wise error for detecting the ice-surface and ice-bottom layers, runs faster and uses approximately 12 million fewer parameters.

Stereo Matching Algorithm Based on Hybrid Convolutional Neural Network and Directional Intensity Difference

Fundamentally, a stereo matching algorithm produces a disparity map or depth map. This map contains valuable information for many applications, such as range estimation, autonomous vehicle navigation and 3D surface reconstruction. The stereo matching process faces various challenges to get an accurate result for example low texture area, repetitive pattern and discontinuity regions. The proposed algorithm must be robust and viable with all of these challenges and is capable to deliver good accuracy. Hence, this article proposes a new stereo matching algorithm based on a hybrid Convolutional Neural Network (CNN) combined with directional intensity differences at the matching cost stage. The proposed algorithm contains a deep learning-based method and a handcrafted method. Then, the bilateral filter is used to aggregate the matching cost volume while preserving the object edges. The Winner-Take-All (WTA) is utilized at the optimization stage which the WTA normalizes the disparity values. At the last stage, a series of refinement processes will be applied to enhance the final disparity map. A standard benchmarking evaluation system from the Middlebury Stereo dataset is used to measure the algorithm performance. This dataset provides images with the characteristics of low texture area, repetitive pattern and discontinuity regions. The average error produced for all pixel regions is 8.51%, while the nonoccluded region is 5.77%. Based on the experimental results, the proposed algorithm produces good accuracy and robustness against the stereo matching challenges. It is also competitive with other published methods and can be used as a complete algorithm

A Comparison of vision-based and CNN-based detector for fish monitoring in complex environment

Aquaculture farming can help soften the environmental impact of overfishing by fulfilling seafood demands with farmed fishes. However, to maintain big scale farms can be challenging, even with the help of underwater cameras affixed in farm cages, because there are hours’ worth of footages to sift through, which can be a laborious task if performed manually. Vision-based system therefore could be deployed to filter useful information from video footages, automatically. This work proposed to solve the above mentioned problems by deploying the; 1) Extended UTAR Aquaculture Farm Fish Monitoring System Framework (UFFMS), being the handcrafted method, and 2) Faster Region Convolutional Neural Network (Faster RCNN), being the CNN-based method, for fish detection. These two methods could extract information about fishes from video footages. Experimental results show that for well-lit footages, Faster RCNN performs better, compared to the extended-UFFMS. However, accuracy of Faster RCNN drops drastically for non-well-lit footages, at an average of 28.57%, despite still having perfect precision scores.

Augmented Skeleton Based Contrastive Action Learning with Momentum LSTM for Unsupervised Action Recognition

Action recognition via 3D skeleton data is an emerging important topic. Most existing methods rely on hand-crafted descriptors to recognize actions, or perform supervised action representation learning with massive labels. In this paper, we for the first time propose a contrastive action learning paradigm named AS-CAL that exploits different augmentations of unlabeled skeleton sequences to learn action representations in an unsupervised manner. Specifically, we first propose to contrast similarity between augmented instances of the input skeleton sequence, which are transformed with multiple novel augmentation strategies, to learn inherent action patterns (“ pattern-invariance ”) in different skeleton transformations. Second, to encourage learning the pattern-invariance with more consistent action representations, we propose a momentum LSTM , which is implemented as the momentum-based moving average of LSTM based query encoder, to encode long-term action dynamics of the key sequence. Third, we introduce a queue to store the encoded keys, which allows flexibly reusing proceeding keys to build a consistent dictionary to facilitate contrastive learning. Last, we propose a novel representation named Contrastive Action Encoding (CAE) to represent human’s action effectively. Empirical evaluations show that our approach significantly outperforms hand-crafted methods by 10–50% Top-1 accuracy, and it can even achieve superior performance to many supervised learning methods (Our codes are available at https://github.com/Mikexu007/AS-CAL ).

Detection of mitotic cells in breast cancer histopathological images using deep versus handcrafted features

One of the most important processes in the diagnosis of breast cancer, which is the leading mortality rate in women, is the detection of the mitosis stage at the cellular level. In literature, many studies have been proposed on the computer-aided diagnosis (CAD) system for detecting mitotic cells in breast cancer histopathological images. In this study, comparative evaluation of conventional and deep learning based feature extraction methods for automatic detection of mitosis in histopathological images are focused. While various handcrafted features are extracted with textural/spatial, statistical and shape-based methods in conventional approach, the convolutional neural network structure proposed on the deep learning approach aims to create an architecture that extracts the features of small cellular structures such as mitotic cells. Mitosis detection/counting is an important process that helps us assess how aggressive or malignant the cancer’s spread is. In the proposed study, approximately 180,000 non-mitotic and 748 mitotic cells are extracted for the evaluations. It is obvious that the classification stage cannot be performed properly due to the imbalanced numbers of mitotic and non-mitotic cells extracted from histopathological images. Hence, the random under-sampling boosting (RUSBoost) method is exploited to overcome this problem. The proposed framework is tested on mitosis detection in breast cancer histopathological images dataset provided from the International Conference on Pattern Recognition (ICPR) 2014 contest. In the results obtained with the deep learning approach, 79.42% recall, 96.78% precision and 86.97% F-measure values are achieved more successfully than handcrafted methods. A client/server-based framework has also been developed as a secondary decision support system for use by pathologists in hospitals. Thus, it is aimed that pathologists will be able to detect mitotic cells in various histopathological images more easily through necessary interfaces.

An adaptive digital stain separation method for deep learning-based automatic cell profile counts

BackgroundQuantifying cells in a defined region of biological tissue is critical for many clinical and preclinical studies, especially in the fields of pathology, toxicology, cancer and behavior. As part of a program to develop accurate, precise and more efficient automatic approaches for quantifying morphometric changes in biological tissue, we have shown that both deep learning-based and hand-crafted algorithms can estimate the total number of histologically stained cells at their maximal profile of focus in Extended Depth of Field (EDF) images. Deep learning-based approaches show accuracy comparable to manual counts on EDF images but significant enhancement in reproducibility, throughput efficiency and reduced error from human factors. However, a majority of the automated counts are designed for single-immunostained tissue sections. New MethodTo expand the automatic counting methods to more complex dual-staining protocols, we developed an adaptive method to separate stain color channels on images from tissue sections stained by a primary immunostain with secondary counterstain. Comparison with Existing MethodsThe proposed method overcomes the limitations of the state-of-the-art stain-separation methods, like the requirement of pure stain color basis as a prerequisite or stain color basis learning on each image. ResultsExperimental results are presented for automatic counts using deep learning-based and hand-crafted algorithms for sections immunostained for neurons (Neu-N) or microglial cells (Iba-1) with cresyl violet counterstain. ConclusionOur findings show more accurate counts by deep learning methods compared to the handcrafted method. Thus, stain-separated images can function as input for automatic deep learning-based quantification methods designed for single-stained tissue sections.