SECA | Perry Enterprises Company LLC

SensorFusion Enhanced Contextual A.I.

SECA is a proprietary concept proposed by Kwasi Perry for context-aware intelligence support that combines sensor fusion, geospatial reasoning, and machine learning to improve interpretation of complex operational environments.

ISRMission Relevance

AIContextual Analysis

GEOINTDomain Foundation

Beyond object detection. Toward operational context.

Modern computer vision can identify objects, but operational value depends on context: where an object is, what surrounds it, what signatures it produces, and how those signals relate to mission conditions.

Capability 01

Geospatial Context

Evaluates location, proximity, infrastructure relationship, environmental setting, and spatial patterns around observed activity.

Capability 02

Sensor-Derived Indicators

Designed around integration of EO, IR/thermal, radar, LiDAR, multispectral, GNSS, INS, and other sensor inputs.

Capability 03

Contextual AI Modeling

Uses machine learning, computer vision, supervised classification, and unsupervised classification to support contextual assessment.

AI Enhanced Air Launched Effects

This earlier white paper explored how artificial intelligence, neural networks, computer vision, machine learning, and unsupervised classification could enhance DoD weapon systems and ISR platforms. It adds a clean technology-lineage layer without crowding the core SECA message.

Autonomous collection: an airborne platform scans terrain while collecting operationally relevant data.

Machine classification: computer vision identifies and classifies terrain features and threat objects.

Operational dissemination: processed information is sent to ground users for near real-time situational awareness.

Designed for future-facing aerospace and ISR environments.

SECA's core concept is adaptable to various platforms and mission sets, but it is particularly well-suited for airborne applications where rapid processing of complex sensor data can provide critical operational insights.

Manned and unmanned aircraft mission support

ISR and situational awareness enhancement

Contextual anomaly and pattern recognition

Research, prototype, test, and evaluation pathway

Operational experience with geospatial intelligence and defense-adjacent work.

Advanced Learning A.I. is led by Kwasi Perry, a former Federal Intelligence Officer with experience spanning geospatial intelligence, deployed operational environments, unmanned systems, and federal research support.

National Geospatial-Intelligence Agency background specializing in GEOINT and IMINT

NSF projects abroad in collaboration with the University of Notre Dame

Environmental Protection Agency geospatial research partner

Experience with defense, research, transportation, and university clients

Vision Transformers, CNNs,and Sensor Fusion.

I am a prospective PhD student focused on the intersection of Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), and Sensor Fusion. My decade of experience as a global technology entrepreneur includes scaling autonomous methodologies converging geo-rectified electroptical sensor data with computer vision for a Federal Highway Administration (FHWA) project and as geospatial intelligence analyst at the National Geospatial-Intelligence Agency (NGA), where I led the deployment of novel sUAS and AI workflows across multi-state jurisdictions, and created geospatial intelligence products for the US Intelligence Community and DoD.

My work has consistently bridged the gap between academic research and industrial application, including technical contributions to multi-institutional NSF projects.

My current research goals involve moving beyond standard convolutional backbones toward hybrid CNN-Transformer architectures.

I am particularly interested in developing hardware-aware AI that optimizes Swin Transformers and MobileViT for real-time, multi-modal sensing on the edge.

By formalizing these methodologies within a doctoral framework, I aim to enhance the global contextual awareness and operational reliability of autonomous systems in critical infrastructure and national security environments.

Technical Inquiries & Collaboration

For research discussions, technical inquiries, collaboration opportunities, or restricted briefing access requests, please contact: