video_analyzer

VideoAnalyzer Class

---

Overview

The VideoAnalyzer class provides tools for loading, analyzing, and processing video files. It supports frame-by-frame analysis using an ImageAnalyzer, saving individual frames, creating timelapse videos, and extracting biological information from video sequences.

---

Class Definition

class VideoAnalyzer:
    def __init__(self):
        """Initialize the VideoAnalyzer."""
        ...

---

Attributes

Attribute	Type	Description
video	cv2.VideoCapture	Video capture object.
current_frame	np.ndarray	Current frame as a NumPy array.
frame_count	int	Total number of frames in the video.
fps	int	Frames per second of the video.
image_analyzer	ImageAnalyzer	ImageAnalyzer instance for image processing.

---

Methods

Initialization

• __init__(self) Initializes a new VideoAnalyzer instance.

---

Video Loading

• load_video(self, video_path) Loads a video file.

  • Parameters:
    • video_path: Path to the video file.

---

Frame Retrieval

• get_frame(self, frame_number=None) -> np.ndarray Retrieves a specific frame or the next frame if no frame number is specified.

  • Parameters:

    • frame_number: Optional frame number to retrieve.

  • Returns: The frame as a NumPy array.

---

Frame Saving

• save_frame(self, output_path, frame=None) Saves the current frame or a specific frame as an image.

  • Parameters:
    • output_path: Path to save the image.
    • frame: Optional frame to save (if not provided, saves the current frame).

---

Frame Analysis

• analyze_frame(self, frame=None) -> Dict[str, Any] Analyzes the current frame or a specific frame using ImageAnalyzer.

  • Parameters:

    • frame: Optional frame to analyze (if not provided, analyzes the current frame).

  • Returns: Analysis results, including cells, nuclei, and mitochondria.

---

Video Analysis

• analyze_video(self, frame_interval=1) -> List[Dict[str, Any]] Analyzes the entire video at specified frame intervals.

  • Parameters:

    • frame_interval: Interval between frames to analyze.

  • Returns: List of analysis results for each analyzed frame.

---

Timelapse Creation

• create_timelapse(self, output_path, start_frame, end_frame, interval=1, resize_factor=1.0) Creates a timelapse video from a range of frames.

  • Parameters:
    • output_path: Path to save the timelapse video.
    • start_frame: Starting frame number.
    • end_frame: Ending frame number.
    • interval: Number of frames to skip between each frame in the timelapse.
    • resize_factor: Factor by which to resize frames (1.0 means no resizing).

---

Cleanup

• close(self) Releases the video capture object.

---

Example Usage

# Initialize VideoAnalyzer
video_analyzer = VideoAnalyzer()

# Load a video file
video_analyzer.load_video("path/to/video.mp4")

# Get a specific frame
frame = video_analyzer.get_frame(100)
print(f"Retrieved frame shape: {frame.shape}")

# Save the current frame
video_analyzer.save_frame("path/to/frame_100.png")

# Analyze a frame
analysis_results = video_analyzer.analyze_frame()
print(f"Analysis results: {analysis_results}")

# Analyze the entire video
video_analysis = video_analyzer.analyze_video(frame_interval=5)
print(f"Video analysis results: {video_analysis}")

# Create a timelapse video
video_analyzer.create_timelapse(
    "path/to/timelapse.mp4",
    start_frame=0,
    end_frame=100,
    interval=2,
    resize_factor=0.5
)

# Close the video capture
video_analyzer.close()

---

Dependencies

• cv2 (OpenCV): For video processing, frame retrieval, and video writing.
• biobridge.tools.image_analyzer.ImageAnalyzer: For image analysis.

---

Error Handling

• The class includes checks for valid input data and handles potential errors during video processing and frame analysis.

---

Notes

• The VideoAnalyzer class is designed for analyzing video sequences, particularly those depicting biological processes.
• It supports frame-by-frame analysis using an ImageAnalyzer to extract biological information such as cells, nuclei, and mitochondria.
• The class provides methods for saving individual frames and creating timelapse videos.
• The analyze_frame method uses an ImageAnalyzer to perform detailed analysis of each frame.
• The analyze_video method allows for analyzing the entire video at specified frame intervals.